a?” ”in?“ 0. RM Mg'
:5... 'h n.
Wmf“""

$.23“; .x‘z" "hf
““145”? 321%?

34)
:6 1'
, , 4: r 5 ~ .
”Emma-- ' .- 3.,“- “ ' 4 ‘ 4-7593:
, . {h ‘ 1. ‘ 38:54:24.! ’
1mm 5-,,
‘73: .39“""
4 ‘ ,
21.5,,
93“”
4. i 4~115m€
| V . ‘ {fir-I"
P'jfi'l'l' .
4:” £339; 3,3 «3:35.-
”4;; .3" 14.1%,?» £323.»,
553, {a 5'95“ '

:‘j’ . ‘ 1 it ‘2. a. H
I | ' n.‘ 93.}: W _:.‘q‘r::é :I’,‘ 3‘ lit-:g‘wz‘ifA’kK'
' r,

‘
“ ,- '::,:.' " 39‘“? 3‘44“"
I ; _ "‘ .2... ask‘ _ ~ficl-J“ - ‘
. ~f r h-$ 3.7%.]? ‘ "‘ if“: .,-. ._gqu£. _
«ow V 74:94:} .‘
. _ J‘ ' 3 {gr-
. .a. u . . .
P X “H .....»..:y, ”‘1‘ ‘. RA v 5 $3433.13,
3'3"},
4 if: ri’fi-J ifig‘

' - 3.. ’ ‘ A“ “J

7: '3'
.
m

‘w.
~14
.

4 «4 . u d, . _
@‘gfi‘fiufig'm ' ' ’ - 4 $§$ 8' u. 7 r . ‘
' '-' 3 . . u? ' - : ~ : - . , .
‘4‘) “ E" - I. Wam- ’ -} 4 . ;. , (a . , 4‘? , .
mi}: 5 4’ '1‘ 1,7 ~- _ - - v 4‘1 {F ”up. 1 '3"! u . ' «r fiiwmx -’
tJ v1} '

4.}912 “£3. ‘ 5;}:
. ’4' w. {11139:

...
.5‘
y-
c n,
..,_ .-

‘al. ‘1
”4 h
“1;.
L54,“
‘1'

345$
’21,

c, .. -
4°43:
1- 4

a... r-
.. :53 7..,.{.>t”"l:. i1 ,
Jana?“ ‘ ‘
my;
1:.“ “’3

,-" 4
“3“ A ~.
“If“: ‘ 29:31: :3 ‘H
“44,-"rfi,

an
{33.14: 4

.1» . , ' ~ '
‘;,§.r...r.,n,:.,, 4. 4.. , 1 _ - g , .- ”. V a» " WWW M
r ”’55"? ‘ ‘ r ' ' ‘ ‘4} 4?“ *5

H.
; g, f '
4_‘;:;.§:

, .
2' #:j-sff‘
.. .214

x ‘ "L". M
«yr. :. z
4", h" "

Q». '

J 34}

1“”.53-

»J)". .ar
4::

.'l'
.
J'err‘v ':

 

. '39-‘18

This is to certify that the

dissertation entitled

Active site studies on Klebsiella aerogenes urease,
a nickel-containing enzyme

presented by

Matthew J. Todd

has been accepted towards fulfillment
of the requirements for

PhD degree in Biochemistry

16% WW

\
Major professor /

Date June 19, 1991

 

MSU is an Affirmative Action/Equal Opportunity Institution ‘ 0-12771

_-——--—__——-‘,-—_A ‘1 —

 

 

ӣ444,444 7

M‘Chigan 4.44444;
Unwersétgv J

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

r

DATE DUE DATE DUE DATE DUE H

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
cWafl-M

 

ACTIVE SITE STUDIES ON KLEBSIELLA AEROGENES UREASE,

A NICKEL-CONTAINING ENZYME

BY

Matthew J. Todd

A DISSERTATION

Submitted to

Michigan State University

in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1991

ABSTRACT

ACTIVE SITE STUDIES ON KLEBSIELLA AEROGENES UREASE,

A NICKEL-CONTAINING ENZYME

BY

Matthew J. Todd

Although ureases play important roles in microbial ni-
trogen metabolism and in the pathogenesis of several human
diseases, little is known of the mechanism of urea hydro-
lysis. As a paradigm for microbial ureases, the enzyme from
Klebsiella aerogenes was studied. This enzyme was purified
1070-fold and shown to consist of three polypeptides (esti—
mated.bb.= 72,000, 11,000, and 9,000) in an apparent oqflnm
quaternary structure, containing 4 mol nickel/mol urease. A
tight-binding inhibitor was exploited to demonstrate the
presence of two mol active site/mol enzyme, thus each cata-
lytic unit ((13272) has two nickel. The heteropolymeric sub-
unit structure was not expected since the well-studied plant
urease from the jack bean is homopolymeric.

K. aerogenes urease is competitively inhibited by
thiols, boronic acids, hydroxamates, phosphate and phos-
phoroamides. UV-visible absorbance studies demonstrated

charge-transfer interactions between thiolate anions and

urease nickel ion, consistent with the presence of at least
one nickel per catalytic unit. Comparison of.K;velues for
several thiolate inhibitors were consistent with electrosta-
tic repulsion of anionic inhibitors by an anionic group on
the enzyme near the thiolate binding site. Acetohydroxamate
(AHA) and phenylphosphorodiamidate (PPD) were slow-binding
competitive inhibitors with ngalues of 2.6 uM and 94 pM,
respectively. Following initial binding to one nickel, the
inhibitors were proposed to form complexes bridging the two
urease active site nickel ions.

The kinetics of K. aerogenes urease thiol modification
were studied using alkylating and disulfide reagents. Reac-
tivity of the essential thiol was affected by substrate and
competitive inhibitors, consistent with a cysteine located
proximal to the active site. The atypical pH dependence for
the rate of inactivation was interpreted as arising from an
interaction between the thiol and a second (as yet unidenti—
fied) ionizing group. The competitive inhibitors phosphate
and PPD (but not AHA) protected one mol thiol/mol catalytic
unit from.modification by 2,2'dithiodipyridine. Using 14C-
iodoacetamide, the essential cysteine was identified as
Cys319, an amino acid conserved among all known urease se-

quences.

to my wife and family

AC KN OWL EDGMENTS

As with any major work, the production of a dissertation
is bound to inconvenience and even burden those around us.
Recognizing this, I would like to acknowledge two individuals
whose sacrifices helped to make it possible. Dr. Hausinger,
not only as mentor, but also as a friend gave me much needed
guidance and then retreated, allowing me to fail or succeed
on my own. Janet, my wife, was willing to sacrifice her own
time and energy, allowing me to pursue these studies. During
most of our marriage, she has endured considerably less at-
tention than she desired, without excess complaint, while
reminding me that there is a "real" world, around which we
should ideally focus our lives.

I would like to thank Scott Mulrooney for providing
pKAUlQ, a plasmid which expresses urease to almost 10% of
the cell protein, Mann Hyung Lee for one preparation of K.
aerogenes urease, Dr. J. Breznek for the use of his scin—
tillation counter, and Dr. J. Leykam for helpful discussions
concerning peptide isolation strategy. Dr. J. Wilson and
Dr. T. Deits gave insightful comments during preparation of
chapters 2-4. The remaining committee members, Dr. C.

Chang, Dr. J. Kaguni, and Dr. D. Lamport were available and

willing to give guidance when needed.

TABLE OF CONTENTS

List of Tables .......................................... viii
List of Figures ........................................... ix
List of Abbreviations .................................... xii

Chapter 1. Introduction

Medical significance .................................. 2
Agricultural significance ............................. 4
Industrial significance ............................... 5
Plant ureases ......................................... 6
Properties of purified urease ......................... 6
Kinetic parameters ................................... 10
Inhibition of urease ................................. 10
Active site residues ................................. 13
References ........................................... 17

Chapter 2. Purification and Characterization of the Nickel-

containing Multicomponent Urease from K. aerogenes

Abstract ............................................. 28
Introduction ......................................... 30
Experimental Procedures .............................. 31
Bacterial growth conditions ..................... 31
Assays .......................................... 32
Preparation of crude extract .................... 32

SDS-polyacrylamide gel electrophoresis .......... 33

Native gel electrophoresis ...................... 34
Native molecular weight determination ........... 34
Stability studies ............................... 35
pH studies ...................................... 35
Amino acid analysis ............................. 36
Nickel analysis ................................. 36
Results .............................................. 37
Urease purification ............................. 37

SDS-polyacrylamide gel electrophoresis of K.
aerogenes urease ........................... 43

Native molecular weight for K. aerogenes urease.43

Attempts to resolve the three urease components.45

Enzyme stability studies ........................ 46
Kinetic parameters .............................. 48
pH studies ...................................... 48
Amino acid analysis ............................. 49
Atomic absorption analysis ...................... 49
Discussion ........................................... 49
References ........................................... 57

Chapter 3. Competitive Inhibitors of Klebsiella aerogenes
Urease: Mechanism of Interaction with the Nickel
Active Site

Abstract ............................................. 60
Introduction ......................................... 62
Experimental Procedures .............................. 63
Materials ....................................... 63
Enzyme purification ............................. 64

Assays .......................................... 64

Spectroscopy .................................... 65

K,- determinations ............................... 66

Inhibitor binding rates in the absence of
substrate .................................. 66

Progress curves in the presence of inhibitor....67

Dissociation rate ............................... 67
Statistical methods ............................. 68
Quantitation of active sites .................... 68
Results .............................................. 69
Thiol inhibition ................................ 69
Spectral interactions of B-ME with urease ....... 72
Phosphate Inhibition ............................ 75
Acetohydroxamate inhibition ..................... 75
PPD inhibition .................................. 84
Quantitation of active sties .................... 85
Other inhibitors ................................ 85
Discussion ........................................... 87
Thiol interactions with urease .................. 89
Phosphate inhibition ............................ 91
Acetohydroxamic acid inhibition ................. 93
Phenylphosphorodiamidate inhibition ............. 96
Boric and boronic acids ......................... 99
References .......................................... 101

Chapter 4. Reactivity of the Essential Thiol of Klebsiella
aerogenes Urease: Effect of pH and Ligands on Thiol
Modification
Abstract ............................................ 104
Introduction ........................................ 106
Experimental Procedures ............................. 107

Enzyme Purification ............................ 107
Kinetics of urease inactivation ................ 108

Reactivation of DTNB- or DTDP-inactivated

urease .................................... 111
Spectroscopic analysis ......................... 111
Curve fitting .................................. 112

Results ............................................. 113

Inactivation of urease by thiol-specific
reagents .................................. 113

Reactivation of urease inactivated by disulfide
reagents .................................. 116

Effect of ligands on the rate of thiol

modification .............................. 119
The effect of pH on leapp ....................... 123
Spectrophotometric quantitation of urease
thiols .................................... 126
Discussion .......................................... 130

Demonstration of an essential thiol in
K. aerogenes urease ....................... 130

Is the essential thiol at the active site? ..... 131

pH dependence of urease inactivation by IAM
and DTDP .................................. 132

Urease inactivation by DTNB .................... 136

Spectroscopic quantitation of urease thiols....l38

References .......................................... 140

(A)
. .

Chapter 5. Identification of the Essential Cysteine Residue
in Klebsiella aerogenes Urease

Abstract ........ - .................................... 142
Introduction ........................................ 143
Experimental Procedures ............................. 146

Enzyme purification ............................ 146

Uniform labelling of cysteine residues
in urease ................................. 146

Specific labelling of the cysteine residues in

native urease ............................. 147
Standardized protocol for peptide mapping ...... 147
Characterization of the specifically labelled

peptide ................................... 148

Results and discussion .............................. 148

Uniform labelling of urease cysteines and de-
velopment of a peptide mapping protocol...148

Specific labelling of the essential cysteine
at pH 6.3 ................................. 152

Specific labelling of the essential cysteine
at high pH ................................ 154

Identification of the cysteine in peptide D....159

Comparison of K. aerogenes urease sequence to
other ureases ............................. 161

References .......................................... 163

Chapter 6. Summary and Future Directions

Urease purification and characterization ............. 165
Urease inhibitors .................................... 169
Active site amino acids .............................. 170

Appendix. Saturation Magnetization of the nickel
centers in urease ................................... 174

Wi

 

~a..¢

..

v...

 

 

LIST OF TABLES

Chapter 2
l. Urease purification from K. aerogenes ............ 4O
2. Stability of K. aerogenes urease ................. 47
3. Amino acid compositions of ureases ............... 51
Chapter 3
1. Thiol Inhibition of K. aerogenes urease .......... 71
Chapter 4
1. Second-order Rate Constants for Loss of

Urease Activity by Thiol-Specific Reagents ..... 115

2. Effect of Urease Ligands on the Rate of
Inactivation by Thiol-Specific Reagents ........ 122

Chapter 5

1. Amino Acid Composition of the Essential
Cysteine-Containing Peptide ................... 160

Wfi

LIST OF FIGURES

Chapter 2

1. DEAE—Sepharose Chromatography of Klebsiella
aerogenes urease ................................ 38

2. Phenyl-Sepharose Chromatography of Klebsiella
aerogenes urease ................................ 39

3. Mono Q Chromatography of Klebsiella aerogenes
urease .......................................... 41

4. Superose 6 Chromatography of Klebsiella
aerogenes urease ................................ 42

5. Sodium dodecyl sulfate-polyacrylamide gel
electrophoresis of several microbial ureases....44

6. The effect of pH on log V /K; for

max

K. aerogenes urease ............................. 50

Chapter 3
1. pH dependence of cysteamine inhibition of urease.70

2. UV-visible spectra of urease in the presence of

thiol ........................................... 73
3. Double-reciprocal plot of l/A absorbance versus

[B-MEJ-l ........................................ 74
4. pH dependence of phosphate inhibition ............ 76

5. Urease progress curves in the presence of AHA....77

6. Rates of AHA binding in the absence of substrate.79

7. Double~reciprocal plot of hmpfl versus [AHA]4....81

8. Progress curves for AHA dissociation ............. 83

9. Quantitation of urease active sites .............. 86

Chapter 4

1. Pseudo—first—order urease activity loss in the
presence of DTDP ............................... 114
2. Saturation of the apparent rate of urease
modification by DTNB: the effects of urea
and borate ..................................... 117
3. Reactivation of DTNB—inactivated urease ......... 118

4. Effect of the competitive inhibitors boric
acid and phosphate on the rate of urease
inactivation by IAM ............................ 120

5. The pH dependence of IAM and DTDP inactivation
of urease ...................................... 124

6. Effect of pH and DTNB concentration on the pseudo-

first-order rate of urease activity loss ....... 125
7. Spectroscopic progress curves for the reaction
of urease with DTDP ............................ 127

8. Activity as a function of urease modification...129

Chapter 5
1. Subunit distribution of urease thiols ........... 149

2. Pro-RPC analysis of uniformly-labelled
trypsin-digested.urease a—subunit ............. 151

3. Effect of phosphate on the modification of
urease thiols by IAM ........................... 153

4. Peptide distribution of urease thiols modified
at low pH ...................................... 155

5. Effect of competitive inhibitors on urease
thiol modification at pH 7.75 .................. 156

6. Peptide distribution of urease thiols modified
at high pH ..................................... 158

7. Comparison of urease sequences surrounding
the essential cysteine ......................... 162

Chapter 6

1. Separation of urease isozymes on Mono-Q ......... 167
2. Pseudo-first order inactivation of urease by
arginine specific reagents ..................... 172

LI ST 0]? ABBREVIATIONS

AHA, acetohydroxamic acid; BBA, benzene boronic acid;
4-Br—BBA, 4—bromobenzene boronic acid, CHES, 2-[N-cyclo-
hexylamino]~ethanesulfonic acid; CAPS, 3-[cyclohexylamino]—
l—propane-sulfonic acid; DTNB, 5,5'-dithiobis(2-nitrobenzoic
acid); DTDP, 2,2'-dithiodipyridine; DTT, dithiothreitol;
EDTA, ethylenediamine tetraacetic acid; Gn-HCl, guanidine
hydrochloride; HEPES N—[2-hydroxyethyl]piperazine-N'-[2-
ethanesulfonic acid]; IAA, iodoacetate; IAM, iodoacetamide;
B-ME, 2-mercaptoethanol; MMTS, methyl methanethiolsulfon—
ate; MBPT, methyl butenyl phosphoric triamide; MES, 2-[N-
morpholino]ethanesulfonic acid; MOPS, 3-[N-morpholino]pro-
panesulfonic acid; NEM, N-ethyl maleimide; PPD, phenylphos—
phorodiamidate; TAPS, N-tris[hydroxymethyl]methyl-3-amino-
propanesulfonic acid; TPCK, L—l—p—tosylamino—Z-pheny1ethyl
chloromethyl ketone; TFA, trifluoroacetic acid; TAPS, N-
tris[hydroxymethyl]methyl—3-aminopropanesulfonic acid;

Tris-HCl, tris(hydroxymethyl)aminomethane~hydrochloride.

mi

INTRODUCTION

Urea is a remarkably stable molecule: the half-life
for the spontaneous degradation of urea is 3.6 years in
aqueous solutions at 38'C (6). Urea is not, however, very
stable in the environment. It is rapidly hydrolyzed to
ammonia and carbamate through the action of the enzyme
urease (EC 3.5.1.5) (5). Carbamate spontaneously degrades
to another molecule of ammonia and carbon dioxide; the car—
bon dioxide is then hydrated to carbonic acid which will
dissociate into H+ and Hcog. Two mol ammonia and one mol
bicarbonate are the net products; thus urea hydrolysis

results in a net increase in pH (Scheme 1).

H30 H20
3 \_ fi \_
H,N-c-NH, .— HgN-C-OH .—_—> 002 \__ H2003 é H00; + H+
+ +
NH; NH;

Scheme 1

Ureases are found in bacteria, algae, filamentous
fungi, yeasts, and plants (reviewed by Mobley and Hausinger;
55). Pathogenic ureolytic microorganisms have been isolated
from humans, where they can colonize the urinary tract, the

colon, or the stomach. Urease activity is also found in

soil, either associated with soil microbes, or as extra-
cellular urease, probably arising from the lysis of soil
microbes or the decomposition of plant tissues (61). The
purpose of this chapter is to describe briefly several areas
where microbial ureases affect the medical and agricultural
community, then to review what is currently known about

urease enzymology.

.Medical significance— Urease is an important virulence
factor in several pathogenic states, especially those asso—
ciated with urinary tract infections. Ureolytic microorgan-
isms in this environment will hydrolyze urea (at 0.4 M in
human urine; 32) and cause a rise in urine pH which can lead
to pyelonephritis (74). Elevated urine pH leads to supersa-
turation of the urine with respect to magnesium and calcium
salts (32) which can precipitate forming stones; 15-20% of
all kidney stones are associated with infections by ureo-
lytic microorganisms (33). The extracellular polysaccharide
of the bacterium is thought to provide an organic matrix
necessary for stone initiation (51) and play a role in ce-
menting the crystalline mineral components together (52).
Growth of urinary calculi imbeds viable bacteria in inter-
stices of the urinary stone (86), inhibiting antibiotic
treatment of such an infection. Precipitation of urine
salts also affects patients with long term urinary catheter—
ization; 86% of catheterized patients have ureolytic bac—

teria present in their urine (56).

In addition to adverse effects caused by the elevation
of pH, ureases release vast quantities of ammonia. Hyper-
ammonemia (83), hepatic encephalopathy (78), and hepatic
coma (76) are conditions caused by elevated levels of am-
monia, which can not be removed from the system by the
liver, either because the liver is damaged or because the
organism is producing an excessive amount of ammonia. Am-
monia is known to have many toxic effects (15), and the
additional ammonia released by ureolytic infections can
contribute to the overall nitrogen burden of the organism.

Treatment of stone-forming kidney infections, hyperam-
monemia, or ammonia toxicity may be more effective using
urease inhibitors; studies using urease inhibitors brflfiv
(85) and in vivo (44, 53, 73) have demonstrated that calculus
formation is reduced in urine lacking urease activity.
Similarly, urease inhibitors have been demonstrated to re-
duce blood ammonia levels in patients with hepatic coma
(83). The rational design of effective inhibitors will re-
quire detailed knowledge of the urease mechanism.

A second type of infection in which urease contributes
to the pathogenicity of a bacterium occurs in the mucosal
lining of the mammalian stomach (23). Colonization of this
hostile environment by the acid sensitive bacterium, Helico-
bacter pylori occurs at intercellular junctions (10). H.
pylori is the etiologic agent of type B gastritis (23, 31)
and has been implicated in the formation of peptic ulcers

(37); eradication of the infection relieves the symptoms

(23). H. pylori possesses an active urease which can ra-
pidly hydrolyze serum urea, producing abnormal amounts of
ammonia, initially thought to create a ”cloud" of neutral pH
immediately surrounding the bacterium (30). Current hypo-
theses to account for H. pylori pathology include: bacterial
secretion of mucin digesting enzymes (80), interference with
the passage of H+ ions from the gastric glands to the lumen
(37), and/or toxicity due to elevated ammonia concentra—
tions, as mentioned above (15, 81). Supporting the role of
urease as a virulence factor of H. pylori, Smoot et al. (81)
demonstrated similar toxicity to human gastric adenocarcin—
oma cells grown with urea and either H. pylori or jack bean
urease. As in the case of urinary tract infections, urease

inhibitors may aid in combatting H. pylori infections.

Agricultural significance- The lack of fixed nitrogen
in the soil is a major limitation to plant growth; there-
fore, fertilizers are used to supplement soil nitrogen
levels when growing commercially important crop plants.
Urea is a good source of nitrogen, because of its low cost,
ease of use, and its high nitrogen content (3). Urea must
be hydrolyzed by soil ureases before it can be assimilated
into plant tissue; uncontrolled hydrolysis can lead to
elevated soil pH, ammonia toxicity, and nitrogen loss as
volatile NH3 (up to 50% for submerged rice crops; 9). The
simultaneous application of urease inhibitors and urea-based

fertilizers could minimize crop damage and enhance the

“HW

efficiency of nitrogen utilization (8, 9). Hydroxamic acids
(69) and phosphoroamide derivatives (8, 16, 50, 61) retard
urea hydrolysis in soils; however, an increase in plant
yield has not yet been demonstrated. One problem regarding
the application of urease inhibitors has been the accumula-
tion of urea in leaves, resulting in leaf-tip necrosis (45).

Obviously, further work in this field is of value.

Industrial significance- Ureases are also of practical
use industrially. Inside a bioreactor, the delivery of nu-
trients to growing cells is diffusion limited; uniform
growth of cells cannot occur if the cells at the site of
nutrient input deplete the medium of nutrients. Attaching
urease to the solid support in the bed of bio-reactors al—
lowed Mak et al. (49) to use urea as a nitrogen source for
growth of a photosynthetic algae, Chlorella emersonii. Re-
actors initiated with immobilized urease had a more uniform
release of ammonia proximal to the growing cells and in-
creased penetration of the gel matrix (48, 49). Ureases
have also been used to remove urea waste from industrial
effluents (G. Prabhakaran, personal communication), or from
blood dialysate in artificial kidneys (67). The waste urea
is hydrolyzed to NH3, which can be easily removed by air

stripping.

Plant ureases- Upon initiation of the current work,

the only extensively studied urease was that from Canavalia

ensiformis, the jack bean plant (reviewed by Blakeley and
Zerner; 6); few microbial ureases had even been purified.
Jack bean urease has the distinction of being the first
enzyme crystallized (84) and the first enzyme shown to pos-
sess nickel (19). Research on the active site of the plant
enzyme allowed Zerner and colleagues to propose a mechanism
for the hydrolysis of urea by urease (22), Scheme 2. In-
deed, many of the initial results on K. aerogenes urease are
interpreted in terms of this model in which each active site
has two nickel ions: one binds to the carbonyl oxygen of
urea, and the other binds a hydroxide for nucleophilic at-
tack on the urea carbon. Elimination of one molecule of
ammonia, followed by release of carbamate and binding of two
molecules of water is thought to regenerate the resting
enzyme. Although plant ureases have been found in seed,
leaf and stem tissues, their role in plant nitrogen metabo-

lism is not well understood (94).

Properties of purified ureases- In contrast to the
homohexameric structure (Mr = 90,770; 88) of jack bean
urease (6) and soybean urease (68), many microbial ureases
have three distinct subunits, a;B,and‘x where a = 60-70 kDa,
and B and y are 8—12 kDa (39, 41, 59, 89), as first descri-
bed for K. aerogenes urease (Chapter 2; 90). The B and 7
subunits have been overlooked when analyzing certain enzymes
by SDS PAGE using gels of less than 12% acrylamide (12, 82).

In one organism, H) pylori, the B and.y subunits have

 

 

 

 

 

 

 

N oanom

\/'
l

 

 

ohmEanookx

 

ONI 00,5
/ kw /

/ ON IN #01:. M fb

 

 

 

 

 

undergone gene fusion into a single polypeptide of 30 kDa
(46). Subunit ratios of urease enzymes have been variable
among species as determined by integrated scanning densito—
metry, a method which assumes that each subunit has a simi-
lar affinity for the dye; therefore definitive comparisons
of subunit ratios are meaningless. Genetic studies have
confirmed the presence of three structural genes in K.
aerogenes (60), P. stuartii (59), P. vulgaris (58), P.
mirabilis (40), and H. pylori (46) and have shown that the
three genes are highly honologous to the single plant gene,
which probably arose via gene fusion. Furthermore, DNA
sequence studies of Scott Mulrooney (60) suggested that
three additional genes are involved in generation of active
urease. One possible function of these additional genes is
in nickel processing (47, 36); deletion of these accessory
genes resulted in production of urease apoenzyme (60).
Nickel has been found in all ureases examined (35).
Calculation of the amount of nickel per catalytic unit re-
quires knowledge of the amount of nickel, and the catalytic
unit molecular weight. Studies incorporating the above in-
formation have shown that the jack bean urease (19, 20) and
the K. aerogenes enzyme (Chapter 2, 3; 90, 91) have 2 mol
nickel/mol catalytic unit. For other bacterial ureases,
0.8—2.1 mol nickel/mol of large subunit has been calculated
(reviewed by Mobley and Hausinger; 55) without determining
the number of catalytic units per enzyme. Biophysical and

spectrosc0pic analysis of the nickel center have only been

done with the jack bean and K. aerogenes enzymes. UV-
visible spectroscopy of K. aerogenes urease in the presence
of thiol inhibitors is described in chapter 3 (91), and
compared to the jack bean enzyme results (4, 18). X-ray
absorption spectroscopy (XAS) of jack bean (1, 14) and K.
aerogenes (Scott and Hausinger, in preparation) ureases have
demonstrated that the nickel is coordinated by a mixture of
nitrogen and oxygen ligands; however the precise identity of
these ligands remains unknown. Lee et al. (47) observed in—
creased reactivity of K. aerogenes apo—urease towards di—
ethylpyrocarbonate at pH 6.5, consistent with (at least
partial) histidine ligation of the nickel. Magnetic suscept-
ibility measurements are consistent with at least a portion
of the nickel being paramagnetic for both K. aerogenes (ap-
pendix) and jack bean (17) ureases. In both cases, diamag—
netism is observed in the presence of thiol compounds, in—
dicating either that the the two nickel ions are antiferro—
magnetically coupled, or that the individual nickel ions
have become low spin (5:0) (17). Low temperature Magnetic
Circular Dichroism (MCD) spectroscopic studies on K.
aerogenes (Michael Johnson, unpublished experiments) and
jack bean (25) ureases show that the two nickel become
antiferromagnetically coupled in the presence of thiols,
suggesting that these inhibitors bridge the two nickel at
the active site. In summary, the active site of urease
includes a novel, poorly characterized bi-nickel

metallocenter.

0"!

A ("1

f1

’1

10

Kinetic parameters- The kinetic parameters of micro—
bial ureases vary moderately and usually reflect the micro-
organisms particular ecological niche. For example, the Kh
values for urea are considerable higher for enzymes purified
from organisms which invade the kidney (2-20 mM; 55) where
the concentration of urea is high (0.4—0.5 M; 32), than from
organisms which colonize the stomach lining (24, 54) where
the concentration of serum urea is 1.5-3.0 mM (54). Urease
Vkax is typically 1000-5000 umol urea hydrolyzed minflmg‘1
(reviewed by Mobley and Hausinger; 55). An important ex-
ception is the urease of U. ureolyticum (V%ax ~ 90,000 umol
min.“1 mg‘l; 82), which is at least 10-fold faster than any
previously purified urease. The urease of this mycoplasma
is thought to create an ammonia and H+ gradient, and the or-
ganism may use this gradient to generate cellular energy
(72). Most ureases have pH optima in the neutral to slightly
basic range (55). Lactobacillus fermentum, however, has a
urease with a pH optimum of 2 (41), which may reflect the

ability of this organism to grow in acidic media.

Inhibition of urease- The inhibition of urease has
many potential applications. For example, derivatives of
hydroxamic acid have been used as chemotherapeutic agents to
reduce blood ammonia levels in patients with hyperammonemia
(83), and to reduce urine ammonia and pH in patients with

urinary infections (33, 62, 73). Serious side effects

ll

accompany the use of these compounds (27), and they have
been suggested to be teratogenic and/or carcinogenic (73).
Phosphoroamide derivatives (which are ~1000—fold more potent
inhibitors) retard urinary stone growth (53, 85) and inhibit
soil urease activity (16, 50, 61). In most instances where
inhibition of urease is desired, the concentration of urea
is present at levels which saturate the enzyme; therefore
an ideal inhibitor would interact with urease by a non-com-
petitive mechanism (57). Many determinations of the type of
inhibition caused by these compounds reported non-competi-
tive (26, 28, 29, 73) or mixed inhibition (28); other
studies compared ISO values (the amount of inhibitor needed
to obtain 50% inhibition of urease activity) without regard
to the mechanism of inhibitor binding (27, 28, 34, 43).
Acetohydroxamic acid and phenylphosphorodiamidate are slow-
binding competitive inhibitors of K. aerogenes urease
(Chapter 3, 91). To generate a useful comparison of the
efficacy of these inhibitors using I50 values, one should
also know the type of inhibition, the K5 and substrate
concentration, and the enzyme concentration (11). Since
these inhbiitors bind and dissociate from urease slowly,
typical kinetic methods (derived assuming equilibrium or
steady state conditions) simply reflect the amount of non-
inhibited enzyme, thus appearing non-competitive (i.e. Vim,
is decreased, K5 is not affected).

The ionized form of hydroxamates are bidentate chela-

tors of metals in solution and are used by some organisms as

12

siderophores (64). Ionized hydroxamates also inhibit metal-
loproteins (ag.thermolysin, 65, 38) through bidentate bind—
ing to the metallocenter. Changes in the UV-visible spectra
of jack bean urease are consistent with their interaction
with the active site nickel ion(s) (21). In addition, UV—
visible spectra of one hydroxamic acid bound to jack bean
urease, trans-cinnamoyl hydroxamate, was consistent with the
ionized form of the hydroxamic acid being present at the
active site, leading Zerner and colleagues to propose that
hydroxamates bind to the urease nickel in a bidentate manner
(21). In contrast, ionized hydroxamic acids do not interact
with horseradish peroxidase; monodentate binding to the ac—
tive site heme is thought to be the cause of hydroxamate in—
hibition of this metalloenzyme (79). The binding of (neu-
tral) acetohydroxamate to K. aerogenes urease is described
in chapter 3 (91), and a distinct mechanism whereby the neu—
tral hydroxamate bridges the two active site nickel (similar
to the transition state of urea hydrolysis, Scheme 2) is

proposed.

Phosphoramides bind to the zinc in metalloproteases
(such as carboxypeptidase A and thermolysin) and inhibit by
acting as transition state analogues (42). Less is known
about the interactions of phosphoramidates with urease.
Zerner and colleagues, studying the reactivation of the jack
bean enzyme inhibited with PPD, MBPT, and phosphoric tri-

amide (2), observed similar reactivation rates with all

13

three compounds. From this observation, derivatives of
phosphoroamide were proposed to be alternative substrates of
jack bean urease; all three being hydrolyzed to a common in-
hibitor, diamidophosphate. The slow-binding of phenylphos-
phorodiamidate to K. aerogenes urease involved saturation
kinetics similar to those seen for AHA (chapter 3; 91). PPD
was also proposed to bind as a transition state analogue.
Although no product analysis was done in either the jack
bean or the bacterial urease study, slowly processed sub-
strates are known to behave kinetically as slow—binding in—
hibitors for several enzymes (57).

In addition to the potent inhibition observed by hy—
droxamates and phosphoramidates, many other compounds are
known to inhibit urease by unknown mechanisms. In chapter 3
-(91), evidence is presented that indicates thiols, phos-
phate, and boric and boronic acids are competitive inhibi—

tors of K. aerogenes urease.

Active site residues- Elucidation of an enzyme
mechanism is not complete without examination of the active
site amino acids which participate in substrate binding,
catalytic turnover, or metal ligation. Chapters 2-5 include
experiments designed to detect these amino acids, including
effects of pH on kinetic parameters, inhibitor and inactiva-
tor studies, and analysis of conserved sequences.
Experiments using the well-studied plant urease demonstrated

a slight increase in Kh upon deprotonation of an ionizable

14

group with pk; = 6.5 (22). Activity (Vhax) required depro-
tonation of the group with pk; = 6.5 while retaining an
enzyme group with pKh = 9.0 in a protonated state. Analo—
gous studies have not been reported for microbial ureases,
except those described in chapter 2 (91), which characterize
the pH dependence of K. aerogenes urease: one enzyme group
(pk; = 6.55) must be deprotonated, and a second enzyme group
(pk; = 8.85) must be protonated for activity (90). The af-
finity of enzyme for substrate is increased slightly below
pH 6.5. While the identity of enzyme groups exhibiting
these pkh's is not known, amino acid specific reagents have
been used to probe their reactivitiy.

Urease activity from plants (66, 71), algae (70) and
bacteria (41, 63, 75, 92) is affected by thiol specific re-
agents. Dixon et al. (22) proposed that jack bean urease
has an essential thiol acting as a general acid during cata—
lytic turnover (Scheme 2). This essential thiol, Cyssgz
(87, 88), was reported to react with modifying reagents only
in the ionized form.with a pKh = 9.0 (66). In contrast, the
essential thiol in K. aerogenes urease interacts with some
nearby amino acid, as yet unidentified, yielding a more com—
plicated pH dependence of inactivation (Chapter 4; 92). The
discrepancy between these two reports may lie in the methods
used to determine the rate of essential thiol modification
(see Chapter 4 for discussion); nevertheless, the analogous
residue has been identified as the essential thiol in bac-

terial urease: Cys319 of the large subunit (Chapter 5; 93).

15

Aside from an essential cysteine, the mechanism of urea
hydrolysis as proposed by Dixon et al., (Scheme 2) invokes
two additional active site amino acids: a carboxylic acid,
and an unidentified base. Evidence supporting an essential
carboxylic acid at the jack bean urease active site is based
on unpublished experiments measuring urease activity in the
presence of triethyloxonium ion. Indirect evidence of an
anionic residue near the active site of bacterial urease
comes from the comparison of various thiol inhibitor K}
values (28, 34, 43, Chapter 3; 91) and comparison of the
second order rate constants for several thiol modifying
reagents (Chapter 4; 92). With regard to the general base,
Sakaguchi et al. (77) have reported inactivation of jack
bean urease by photo-oxidation of histidine residues using
methylene blue; urease inhibitors prevented this inactiva-
tion. Lee et al., (47) have seen loss of K. aerogenes
urease activity using diethylpyrocarbonate, a modifying
reagent which reacts with histidine residues.

Comparison of published urease sequences from K.
aerogenes (60), jack bean (88), P mirabilis (40), P.
vulgaris (58), U. ureolyticum (7), and H. pyloris (46)
demonstrate that ten histidine are conserved among these
diversely represented species. One of these histidine resi-
dues may correspond to the general base described above. In
addition, several of these histidines may serve as ligands

to the nickel, as described above.

16

The following chapters describe my studies on K.
aerogenes urease. First I purify the nickel-containing
enzyme 1070-fold and show that this and several other micro—
bial ureases are heteropolymeric (Chapter 2; published in J.
Biol. Chem., 262, 5963-5967). I then demonstrate that some
competitive inhibitors directly interact with the nickel
center, and propose models for the interaction of these
inhibitors with the bi-metal active site (Chapter 3;
published in J. Biol. Chem., 264, 15836—15842). Finally, I
describe the unusual reactivity of the essential cysteine in
urease (Chapter 4; published in J. Biol. Chem., 266, 10260-
10267), and I identify this amino acid as Cys319 in the a

subunit (Chapter 5; 93).

REFERENCES

Alagna, L., Hasnain, S.S., Piggot, B., & Williams, D.
J. (1984) The nickel ion environment in jack bean
urease. Biochem. J. 220, 591-595

Andrews, R. K., Dexter, A., Blakeley, R. L., & Zerner,
B. (1986) Jack bean urease (EC 3.5.1.5). VIII. On
the inhibition of urease by amides and esters of
phosphoric acid. J. Am. Chem. Soc. 108, 7124-7125

Beaton, J. D. (1978) Urea: its popularity grows as a
dry source of nitrogen. Crops Soils 30, 11-14

Blakeley, R. L., Dixon, N. E., & Zerner, B. (1983)
Jack bean urease VII. Light scattering and Ni (II)
spectrum. Thiolate-nickel (II) charge-transfer peaks
in the spectrum of the B-mercaptoethanol-urease
complex. Biochim. et Biophys. Acta 774, 219-229

Blakeley, R. L., Hinds, J. A., Kunze, H. B., Webb, E.
C., & Zerner, B. (1969) Jack bean urease (EC
3.5.1.5). Demonstration of a carbamoyl-transfer
reaction and inhibition by hydroxamic acids. Biochem.
8. 1991—2000

Blakeley, R. L., & Zerner, B. (1984) Jack bean
urease: The first nickel enzyme. J..Mol. Catalysis
23, 263-292

Blanchard, A. (1990) Ureaplasma urealyticum urease
genes; use of a USA tryptophan codon. Mbl..Microbiol.
4. 669—676

Bremner, J. M., & Krogmeier, M. J. (1988) Elimination
of the adverse effects of urea fertilizer on seed
germination, seedling growth, and early plant growth in
soil. Proc. Natl. Acad. Sci. 85, 4601-4604

17

10.

11.

12.

13.

14.

15.

16.

18

Byrnes, B. H., Savant, N. K., & Craswell, E. T. (1983)
Effect of a urease inhibitor, phenylphosphorodiamide,
on the efficience of urea applied to rice. Soil Sci.
Soc. Am. J. 47, 270-274

Chen, X. G., Correa, P., Offerhaus, J., Rodriguez, E.,
Janney, P., Hoffmann, E., Fox, J., Hunter, F., &
Diavolitsis, S. (1986) Ultrastructure of the gastric
mucosa harboring campylobacter-like organisms. Am. J.
of Clin. Path. 86, 575-582

Cheng, Y. C., & Prusoff, W. H. (1973) Relationship
between the inhibition constant (Kfl and the
concentration of inhibitor which causes 50 percent
inhibition (Imfl of an enzymatic reaciton. Biochem.
Pharmacol. 22, 3099-3108

Christians, 5., & Kaltwasser, H. (1986) Nickel-
content of urease from Bacillus pasteurii. Arch.
.Microbiol. 145, 51-55

Clark, P. A., & Wilcox, D. E. (1989) Magnetic
properties of the nickel enzymes urease, nickel-
substituted carboxypeptidase, A, and nickel-substituted
carbonic anhydrase. Inorg. Chem. 28, 1326

Clark, P. A., Wilcox, D. E., & Scott, R. A. (1990) X—
ray absorption spectroscopic evidence for binding of
the competitive inhibitor 2-mercaptoethanol to the
nickel sites of jack bean urease. A new Ni-Ni
interaction in the inhibited enzyme. Inorg. Chem. 29,
579-581

Cooper, A. J., & Plum, F. (1987) Biochemistry and
Physiology of Brain Ammonia. Phys. Rev. 67, 440-519.

Creason, G. L., Schmitt, M. R., Douglass, E. A., &
Hendrickson, L. L. (1990) Urease inhibitory activity
associated with N—(n-butyl)thiophosphoric triamide is
due to formation of its oxon analog. Soil Biol.
Biochem. 22, 209-211

Ad.

a

s

17.

18.

19.

20.

21.

22.

23.

24.

19

Day, E. P., Peterson, J., Todd, M. J., & Hausinger, R.
P. (1989) Saturation magnetization of the nickel
centers in urease. J. Inorg. Biochem. 36, 305

Dixon, N. E., Blakeley, R. L., & Zerner, B. (1980a)
Jack bean urease (EC 3.5.1.5). III. The involvement of
active-site nickel ion in inhibition by B-mercapto-
ethanol, phosphoramidate, and fluoride. Can. J.
Biochem. 58, 481-488

Dixon, N. E., Gazzola, C., Blakeley, R. L., & Zerner,
B. (1975a) Jack bean urease (EC 3.5.1.5). A
metalloenzyme. A simple biological role for nickel?
J. Am. Chem. Soc. 97, 4131-4133

Dixon, N. E., Gazzola, C., Watters, J. J., Blakeley, R.
L., & Zerner, B. (1975b) Inhibition of jack bean
urease (EC 3.5.1.5) by acetohydroxamic acid and by
phosphoramidate. An equivalent weight for urease. J.
Am. Chem. Soc. 97, 4130-4131

Dixon, N. E., Hinds, J. A., Fihelly, A. K., Gazzola,
C., Winzor, D. J., Blakeley, R. L., & Zerner, B.
(1980b) Jack bean urease (EC 3.5.1.5). IV. The
molecular size and the mechanism of inhibition by
hydroxamic acids. Spectrophotometric titration of
enzymes with reversible inhibitors. Can. J. Biochem.
58. 1323-1334

Dixon, N. E., Riddles, P. W., Gazzola, C., Blakeley, R.
L., & Zerner, B. (1980c) Jack bean urease (EC .
3.5.1.5). V. On the mechanism of action of urease on
urea, formamide, acetamide, N-methylurea, and related
compounds. Can. J. Biochem 58, 1335-1344

Dooley, C. P., & Cohen, H. (1988) The clinical
significance of Campylobacter pylori. Annals of Int.
.Med. 108, 70-79

Dunn, B. E., Campbell, G. P., Perez-Perez, G. I., &
Blaser, M. J. (1990) Purification and character-
ization of urease from Helicobacter pylori. J. Biol.
Chem. 265, 9464-9469

25.

26.

27.

28.

29.

30.

31.

32.

33.

20

Finnegan, M. G., Kowel, A. T., Werth, M. T., Clark, P.
A., Wilcox, D. E., & Johnson, M. K. (1991) Variable-
temperature magnetic circular dichroism spectroscopy as
a probe of the electronic and magnetic properties of
nickel in jack bean urease. J. Am. Chem. Soc. 113,
4030-4032

Fishbein, W. N., & Carbone, P. P. (1965) Urease
catalysis. II. Inhibition of the enzyme by
hydroxyurea, hydroxylamine, and acetohydroxamic acid.
J. Biol. Chem. 240, 2407-2414

Fishbein, W. N. (1982) Hydroxamic acids as urease
inhibitors for medical and veterinary use. Chemistry
and biology of hydroxamic acids (H. Kehl, ed), pp. 94-
103. Karger, Basel

Gale, G. R. & Atkins, L. M. (1969) Inhibition of
urease by hydroxamic acids. Arch. int. Pharmacodyn.
180. 289-298

Gatermann, S. John, J., & Marre, R. (1989) Staphylo-
coccus saprophyticus urease: characterization and
contribution to uropathogenicity in unobstructed
urinary tract infection of rats. Infect. Immun. 57,
110-116

Goodwin, C. 8., Armstrong, J. A., Marshall, & B. J.
(1986) Campylobacter pyloridis, gastritis, and peptic
ulceration. J. Clin. Path. 39, 353-365

Graham, D. Y., & Klein, P, D. (1987) Campylobacter
pyloridis gastritis: the past, the present, and
speculations about the future. Am. J. Gastroenter.
82. 283-286

Griffith, D. P., Musher, D. M., & Hin, C. (1976)
Urease. The primary cause of infection-induced urinary
stones. Invest. urol. 13, 346-350

Griffith, D. P. (1978) Struvite stones. Kidney int.
13' 372‘82

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

21

Hase, J. & Kobashi, K. (1967) Inhibition of Proteus
vulgaris urease by hydroxamic acids. J. Biochem. 62,
293-299

Hausinger, R. P. (1987) Nickel utilization by
microorganisms. Microbiol. Rev. 51, 22-42

Hausinger, R. P. (1990) Mechanisms of metal ion
incorporation into metalloproteins. Biofactors 2,
179-184

Hazell, S. L., & Lee, A. (1986) Campylobacter
pyloridis, urease, hydrogen ion back diffusion, and
gastric ulcers. Lancet ii 15-17

Holmes, M. A., & Matthews, B. W. (1981) Binding of
hydroxamic acid inhibitors to crystalline thermolysin
suggests a pentacoordinate zinc intermediate in
catalysis. Biochem. 20, 6912-6920

Hu, L., Nicholson, E. B., Jones, B. D., Lynch, M. J., &
Mobley, H. L. T. (1990) .Mbrganella morganii urease:
purification, characteization, and isolation of gene

(sequences. J. Bacteriol. 172, 3073-3080

Jones, E. D. & Mobley, H. L. T. (1988) Proteus
mirabilis urease: nucleotide sequence determination
and comparison with jack bean urease. J. Bacteriol.
171. 6414-6422

Kakimoto, S., Sumino, Y., Kawahara, K., Yamazaki, E., &
Nakatsui, I. (1990) Purification and characterization
of acid urease from Lactobacillus fermentum. Appl.
.Microbiol. Biotechnol. 32, 538-543

Kam, C. M., Nishino, N., & Powers, J. C. (1979)
Inhibition of thermolysin and carboxypeptidase A by
phosphoramidates. Biochem. 18. 3032-3038

Kobashi, K., Kumaki, K., & Hase, J. (1971) Effect of
acyl residues of hydroxamic acids on urease inhibition.
Biochim. Biophys. Acta. 227, 429-441

44.

45.

46.

47.

48.

49.

50.

51.

52.

22

Kobashi, K., Munakata, K., Takebe, S., & Hase, J.
(1980) Therapy for urolithiasis by hydroxamic acids.
II. Urease inhibitory potency and urinary excretion
rate of hippurohydroxamic acid derivatives. J. Pharm.
Dyn. 3. 444-450

Krogmeier, M. J., McCarty, G. W., & Bremner, J. M.
(1989) Potential phytotoxicity associated with the use
of soil urease inhibitors. Proc. Natl. Acad. Sci. 86,
1110-1112

Labigne, A., Cussac, V., & Courcous, P. (1991)
Shuttle cloning and nucleotide sequences of
Helicobacter pylori genes responsible for urease
activity. J. Bacteriol 173, 1920-1931

Lee, M. Y., Mulrooney, S. B., & Hausinger, R. P.
(1989) Purification, characterization, and in-vivo
reconstitution of Klebsiella aerogenes urease
apoenzyme. J. Bacteriol. 172, 4427-4431

Mak, A. L., & Trevan, M. D. (1988) Urea as a nitrogen
source for calcium-alginate immobilized Chlorella.
Enzyme Microb. Tbchnol. 10, 207—213

Mak, A. L., Trevan, M. D., Rogers, L. J., & Gallon, J.
R. (1989) Urea as a nitrogen source for immobilized
Chlorella. Biochemistry of the algae and cyanobac-
teria. Proc. Phytochem. Soc. Eur. 1989, 362-363

Martens, D. A., & Bremner, J. M. (1984) Effectiveness
of phosphoroamides for retardation of urea hydrolysis
in soils. Soil. Sci. Soc. Am. J. 48, 302-305

McLean, R. J. C., Nickel, J. C., Beveridge, T. J., &
Costerton, J. W. (1989) Observations of the ultra-
structure of infected kidney stones. J..Med.
Microbiol. 29, 1-7

McLean, R. J. C., Nickel, J. C., Cheng, K. J., &
Costerton, J. W. (1988) The ecology and pathogenicity
of urease producing bacteria in the urinary tract. CRC
Crit. Rev..Microbiol. 16, 37-79

r!»
.4v

wa
.s'

744
.s.

53.

54.

55.

56.

57.

58.

59.

60.

61.

23

Millner, O. E., Andersen, J. A., Appler, M. E.,
Benjamin, C. E., Edwards, J. G., Humphrey, D. T., &
Shearer, E. M. (1982) Flurofamide: a potent
inhibitor of bacterial urease with potential clinical
utility in the treatment of infection induced urinary
stones. J. urol. 127, 346—350

Mobley, H. L. T., Cortesia, M. J., Rosenthal, L. E., &
Jones, B. D. (1988) Characterization of urease from
Campylabacter pylori. J. Clin. Microbiol. 26, 831-836

Mobley, H. L. T., & Hausinger, R. P. (1989) Microbial
ureases: significance, regulation, and molecular
characterization. .Microbial. Rev. 53, 85-108

Mobley, H. L. T., & Warren, J. W. (1987) Urease-
positive bacteriuria and obstruction of long-term
urinary catheters. J. Clin. Microbiol. 25, 2216-2217

Morrison, J. F. & Walsh, C. T. (1988) The behavior and
significance of slow-binding enzyme inhibitors. Adv.
Enzymal. 61, 201-301

Morsdorf, G. & Kaltwasser, H. (1990) Cloning of the
genes encoding urease from Proteus vulgaris and
sequencing of the structural genes. FEMS Microbiol.
Lett. 66, 67-74

Mulrooney, S. B., Lynch, M. J., Mobley, H. L. T., &
Hausinger, R. P. (1988) Purification,
characterization, and genetic organization of
recombinant Providencia stuartii urease expressed in
Eschericia cali. J. Bacterial. 170, 2202-2207

Mulrooney, S. B., & Hausinger, R. P. (1990) Sequence
of the Klebsiella aerogenes urease genes and evidence
for accessory proteins facilitating nickel
incorporation. J. Bacterial. 172, 5837-5843

Mulvaney, R. L., & Bremner, J. M. (1981) Control of
urea transformations in soils. Sail Biochemistry, vol.
5. (E. A. Pau and J. N. Ladd, eds.) pp. 153-196,
Marcel Dekker, Inc., New York

62.

63.

64.

65.

66.

67.

68.

69.

70.

24

Munakata, K., Kobashi, K., Takebe, S., & Hase, J.
(1980) Therapy for urolithiasis by hydroxamic acids.
III. urease inhibitory potency and urinary excretion
rate of N-acylglycinohydroxamic acids. J. Pharm. Dyn.
3. 451-456

Nakano, H., Takenishi, S., & Watanabe, Y. (1984)
Purification and properties of urease from
Brevibacterium ammoniagenes. Agric. Biol. Chem. 48,
1495-1502

Nielands, J. B. (1967) Hydroxamic acids in Nature.
Science 156, 1443-1447

Nishino, N. & Powers, J. C. (1979) Design of potent
reversible inhibitors for thermolysin. Peptides
containing zinc coordinating ligands and their use in
affinity chromatography. Biochem. 18, 4340-4347

Norris, R., & Brocklehurst, K. (1976) A convenient
method of preparation of high-activity urease from
Canavalia ensifarmis by covalent chromatography and an
investigation of its thiol groups with 2,2'-dipyridyl
disulphide as a thiol titrant and reactivity probe.
Biochem. J. 159. 245-257

Piskin, E., & Chang, T. M. S. (1981) Urea and ammonia
removal from dialysate in an oxygenator: effects of
dialysate and air flow rates. J..Membrane Sci. 9,
343-350

Polacco, J. C., & Havir, E. A. (1979) Comparisons of
soybean urease isolated from seed and tissue culture.
J. Biol. Chem 254 1707-1715

Pugh, K. B., & Waid, J. S. (1969) The influence of
hydroxamates on ammonia loss from various soils treated
with urea. Soil Biol. Biochem. 1, 207-217

Rai, A. K., & Singh, S. (1987) Urease of blue-green
algae (cyanobacteria) Anabene dalialum and Anacystis
nidulans. Curr. Microbiol. 16, 113-117

4. u

p
N.

Nu

-. k9

71.

72.

73.

74.

75.

76.

77.

78.

79.

25

Riddles, P. W., Andrews, R. K., Blakeley, R. L., &
Zerner, B. (1983) Jack bean urease. VI.
Determination of thiol and disulfide content.
Reversible inactivation of the enzyme by the blocking
of the unique cysteine residue. Biochim. Biophys.
Acta. 743. 115-120

Romano, N., Tolone, G., Ajello, F., & LaLicata, R.
(1980) Adenosine 5'-triphosphate synthesis induced by
urea hydrolysis in Ureaplasma urealyticum. J.
Bacteriol. 144, 830-832

Rosenstein, I. J. M., & Hamilton-Miller, J. M. T.
(1984) Inhibitors of urease as chemotherapeutic
agents. CRC Crit. Rev. Microbial. 11, 1-12

Rubin, R. H., Tolkoff—Rubin, N. E., & Cotran, R. S.
(1986) Urinary tract infection, pyelonephritis, and
reflux nephropathy. The Kidney. (B. M. Bremner & F.
C. Rector, eds) pp. 1085-1141. W. B. Saunders Co.,
Philadelphia

Saada, A. B., & Kahane, I. (1988) Purification and
characterization of urease from ureaplasma urealyticum.
Zbl. Bakt. Hyg. A 269, 160-167

Sabbaj, J., Sutter, V. L., & Finegold, S. M. (1970)
Urease and deaminase activities of fecal bacteria in
hepatic coma. Antimicrob. Agents Chemotherapy. 1970,
181-185

Sakaguchi, K., Mitsui, K., Kobashi, K., & Hase, J.
(1983) Photo-oxidation of jack bean urease in the
presence of methylene blue. J. Biochem. 93, 681-686

Samtoy, B. & DeBeukelar, M. M. (1980) Ammonia
encephalopathy secondary to urinary tract infection
with Proteus mirabilis. Pediatrics 65. 294-297

Schonbaum, G. R. (1973) New complexes of peroxidases
with hydroxamic acids, hydrazides, and amides. J.
Biol. Chem. 248, 502-511

.IV

«(A u

shy

80.

81.

82.

83.

84.

85.

86.

87.

88.

26

Slomiany, B. L., Bilski, J., & Murty, U. L. (1987)
Campylabacter pyloridis degrades mucin and undermines
gastric mucosal integrity. Gastraenteralogy 92, 1645

Smoot, D. T., Mobley, H. L. T., Chippendale, G. R.,
Lewison, J. F., & Resau, F. H. (1990) Helicobacter
pylori urease activity is toxic to human gastric

epithelial cells. Infec. and Immun. 58, 1992-1994

Stemke, G. W., Robertson, J. A., & Nhan, M. (1987)
Purification of urease from Ureaplasma urealyticum.
Can. J..Micrabiol. 33, 857-862

Summerskill, J., Thorsell, E., Feinberg, J. H., &
Aldrete, J. S. (1967) Effects of urease inhibition in
hyperammonemia: clinical and experimental studies with
acetohydroxamic acid. Gastraent. 54, 20-26

Sumner, J. B. (1926) The isolation and crystallization
of the enzyme urease. J. Biol. Chem. 69. 435-441

Takebe, S., Numata, A., & Kobashi, K. (1984) Stone
formation by Ureaplasma urealyticum in human urine and
its prevention by urease inhibitors. J. Clin.
.Microbiol. 20, 869-873

Takeuchi, H., Takayama, H., Konishi, T., & Tomoyoshi,
T. (1984) Scanning electron microscopy detects
bacteria within infection stones. J. Urol. 132, 67-69

Takishima, K., Mamiya, G., & Hata, M. (1983) Amino
acid sequence of a peptide containing an essential
cysteine residue of jack bean urease. Frontiers in
biochemical and biophysical studies of proteins and
membranes (T. Y. Liu, S. Sakakibara, A. N. Schecter,
K. Yagi, H. Yajima, and K. T. Yasunobu, eds) Elsevier,
N.Y.

Takishima, K., Suga, T., & Mamiya, G. (1988) The
structure of jack bean urease. The complete amino acid
sequence, limited proteolysis and reactive cysteine
residues. Eur. J. Biochem. 175, 151-165

89.

90.

91.

92.

93.

94.

27

Thirkell, D., Myles, A. D., Precious, B. L., Frost, J.
S., Woodall, J. C., Burdon, M. G., & Russell, W. C.
(1989) The urease of Ureaplasma urealyticum. J. Gen.
Microbiol. 135. 315-323

Todd, M. J., & Hausinger, R. P. (1987) Purification
and characterization of the nickel-containing
multicomponent urease from Klebsiella aerogenes. J.
Biol. Chem. 262, 5963-5967

Todd, M. J., & Hausinger, R. P. (1989) Competitive
inhibitors of Klebsiella aerogenes urease. J. Biol.
Chem. 264. 15835-15842

Todd, M. J., & Hausinger, R. P. (1991a) Reactivitiy
of the essential thiol of Klebsiella aerogenes urease.
Effect of pH and ligands on thiol modification. J.
Biol. Chem. 266, 10260-10267

Todd, M. J., & Hausinger, R. P. (1991b)
Identification of the essential cysteine residue in
Klebsiella aerogenes urease. J. Biol. Chem.
submitted

Winkler, R. G., Blevins, D. G., Polacco, J. C., &
Randall, D. D. (1988) Ureide catabolism in nitrogen
fixing legumes. Trends Biochem. Sci. 13, 97-100

PURIFICATION AND CHARACTERIZATION OF THE NICKEL-CONTAINING

HULTICOKPONENT UREASE FROM KLEBSIELLA AEROGENES

ABSTRACT

Klebsiella aerogenes urease was purified 1,070-fold
with a 25% yield by a simple procedure involving DEAE-
Sepharose, phenyl-Sepharose, Mono Q, and Superose 6 chrom—
atographies. The enzyme preparation was comprised of three
polypeptides with estimated M1. = 72,000, 11,000, and 9,000
in a “25474 quaternary structure. The three components re-
mained associated during native gel electrophoresis, Mono Q
chromatography, and Superose 6 chromatography despite the
presence of thiols, glycols, detergents, and varied buffer
conditions. The apparent compositional complexity of K.
aerogenes urease contrasts with the simple well-character-
ized homohexameric structure for jack bean urease (Dixon, N.
E., Hinds, J. A., Fihelly, A. K., Gazzola, C., Winzor, D.
J., Blakeley, R. L., and Zerner, B. (1980) Can. J. Biochem.
58, 1323-1334); however, heteromeric subunit compositions
were also observed for the enzymes from Proteus mirabilis,
Sparosarcina ureae, and Selenamanas ruminantium. K.
aerogenes urease exhibited a K5 for urea of 2.8 i 0.6 mM
and a Vhaxiof 2,800 i 200 umol of urea min’l mg‘l at 37'C in

25 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid,

28

29

5.0 mM EDTA buffer, pH 7.75. The enzyme activity was stable
in 1% sodium dodecyl sulfate, 5% Triton X-100, 1 M KCl, and
over a pH range from 5 to 10.5, with maximum activity ob-
served at pH 7.75. Two active site groups were defined by
their pKQ values of 6.55 and 8.85. The amino acid composi-
tion of K. aerogenes urease more closely resembled that for
the enzyme from Brevibacter ammoniagenes (Nakano, H.,
Takenishi, S., and Watanabe, Y. (1984) Agric. Biol. Chem

48, 1495-1502 than those for plant ureases. Atomic absorp—
tion analysis was used to establish the presence of 2.1 i
0.3 mol nickel/mol 72,000-dalton subunit in K. aerogenes

urease .

‘ !

VPL

«an

 

s\k

Introduction

Urease is a nickel-containing enzyme which catalyzes
the hydrolysis of urea to form carbon dioxide and ammonia
(1,2). The archetype urease, isolated from jack bean, has
been intensively studied since 1926 when Sumner first
crystallized this protein (3) . Jack bean urease (M1. =
590,000) is a hexamer of identical subunits (4) of known
amino acid sequence (AL.= 90,790) (5). This plant enzyme
contains two nickel ions per subunit (6), and the metal ions
are apparently coordinated by oxygen and nitrogen ligands
(7). Whereas soybean urease appears to be analogous in
size, structure, and nickel content (8), microbial ureases
are distinct from the jack bean enzyme. For example,
ureases from Brevibacter ammoniagenes and Bacillus pasteurii
are smaller in subunit size (Afi.= 67,000 and 65,000) and
native size (Afi.= 200,000 and 230,000) and possess a single
nickel ion per subunit (9,10). The Selenamanas ruminantium
urease is also smaller (native hfi.= 360,000), but like the
plant enzymes it contains two nickel ions per subunit (hfi.=
70,000) (11). Active site differences may also exist in the
microbial and plant enzymes as shown by their differences in
susceptibility to various inhibitors (12).

Herein we extend the comparison of plant and microbial

ureases by characterizing the isolated Klebsiella aerogenes

3O

31

enzyme. Previously, Friedrich and Magasanik (13) purified
K. aerogenes urease 24-fold and examined the cellular re-
gulation of this enzyme. In addition, Kamel and Hamed (14)
described several properties of 150-fold-purified urease
from the related organism, Aerabacter aerogenes PRL—R3. In
this report, the nickel-containing K. aerogenes urease is
intensively characterized and shown to have a novel urease

structure comprised of three distinct polypeptide chains.

EXPERIMENTAL PROCEDURES

Bacterial growth conditions- Klebsiella aerogenes
CG253, obtained from Boris Magasanik and Alexander Ninfa
(Massachusetts Institute of Technology), was cultured at
37’C in minimal media which contained 10 g sucrose, 10.5 g
K¢HPO,, 4.5 g KHZPO4, 0.2 g MgSOz-7Hy3, 2.5 ml trace min-
erals, and 0.24 g urea (filter sterilized) per liter. The
trace mineral solution contained, per 100 ml, 20 mg
ZnSO4o7HéO, 10 mg H3BO3, 10 mg NaSeO3, 10 mg NaMoO4-2H¢O, 10
mg NiC12-6H20, 5 mg CuSO4-5HZO, 2 mg AlK(SO4)2-12HZO and 1 ml
H2804 (15). Large scale aerobic batch culture was performed
by using a New Brunswick 25L Microferm fermenter with rapid
agitation. The culture was grown to an absorbance (600 nm)
of 1.8 and harvested by using a Pellicon Cassette system
concentrator (Millipore). The cells (80 to 108 g wet weight

per 22 L) were washed with 20 mM potassium phosphate, 1 mM

32

EDTA, 1 mM B-mercaptoethanol, pH 7.0 (abbreviated PEB)

buffer and either used immediately or stored at -20’C.

Assays- Urease was assayed by measuring the rate of
release of ammonia from urea. The released ammonia was
converted to indophenol whose absorbance was monitored at
625 nm (16). Specific activity of urease was defined as
umol urea hydrolyzed min‘lmg“l at 37'C in 25 mM HEPES (N-2-
hydroxyethylpiperazine-N'-2-ethane sulfonic acid), 5.1 mM
EDTA, pH 7.75 buffer containing 25 mM urea. Protein was
assayed as described by Lowry et al. (17) with bovine serum
albumin as the standard. In addition, the absorbance at 280
nm was used to quantitate highly purified samples of urease.
An extinction coefficient of 8.5 at 280 nm was obtained for
a 1% solution of the isolated enzyme, based on the protein
concentration obtained by using the Lowry method. All
UV/visible absorbance determinations were made by using a

Gilford Response spectrophotometer.

Preparation of crude extract- The cells were suspended
in an equal volume of PEB buffer containing 1.0 mM toluene
sulfonyl fluoride and disrupted by two passages through a
French pressure cell at 16,000 psi. In some cases, other
protease inhibitors were used including bestatin, pep-
statin, leupeptin, aprotinin, and phosphoramidone. Mem-
branes and cellular debris were removed by centrifugation at

10,000 x g for 60 min at 4‘C. All isolation procedures,

33

except that using the Fast Protein Liquid Chromatography
(FPLC) system, were carried out at 4°C and using PEB buffer

with the indicated additions.

SDS-palyacrylamide gel electrophoresis- The molecular
weight for the urease polypeptides and the purity of samples
were assessed by using SDS-polyacrylamide gel electro-
phoresis in a Hoefer SE 600 electrophoresis apparatus as
described by Laemmli (18). The samples were denatured for 5
min at 100'C in 0.0625 M Tris buffer, pH 6.8 containing 3%
SDS, 10% glycerol, and 5% B-mercaptoethanol. These samples
were then electrophoresed at 30 mA constant current through
a 3% or 4.5% acrylamide stacking gel and a 10%, 12% or 20%
acrylamide running gel, each 1.5 mm in thickness. To im-
prove molecular weight estimates for small polypeptides, 20%
acrylamide gels possessed a 200:1 ratio of acrylamide to
methylene bisacrylamide. In some cases, 5% to 15% acryl-
amide gradient gels were prepared to enhance the resolution
of the entire range of peptides. Standards (Bio-Rad, Cal-
biochem, Sigma) used for comparison were: myosin, hfi.=
200,000; B-galactosidase, Mr = 116,250; phosphorylase b, Mr
= 92,500; jack bean urease, Afi.= 90,790 (5); bovine serum
albumin, Mr = 66,200; ovalbumin, Mr 2 45,000; carbonic
anhydrase, Mr = 31,000; chymotrypsinogen, Mr = 25,000;
soybean trypsin inhibitor, Mr = 21,500; myoglobin, Mr =
16,950; a-lactalbumin, Mr = 14,200; lysozyme, M1. = 14,000;

cytochrome C, hfi.= 12,500; and myoglobin cyanogen bromide

34

fragments, Nfi.= 14,000, 8,160, 6,210, and 2,510. Gels were
stained with Coomassie Brilliant blue or Silver stain (19)
to detect proteins and with dansyl hydrazine to detect

glycoproteins (20).

Native gel electrophoresis— Nondenaturing gel electro—
phoresis was carried out by using the buffers described by
Laemmli (18) without SDS. Samples were electrophoresed at
35 mA constant current into the 3% acrylamide stacking gel
and 5%, 7%, or 9% acrylamide running gel, 1.5 mm in thick-
ness. Urease was detected in the gels either by a phenol
red activity stain analogous to the cresol green stain
described by Blattler et al. (21) or by the nitroblue
tetrazolium activity stain described by Fishbein (22).

Other proteins were detected by Coomassie brilliant blue or

Silver stain methods.

Native molecular weight determination- The native
molecular weight of K. aerogenes urease was estimated by two
methods. Native gel electrophoresis was carried out on
samples and standards by using several gels of varied acryl-
amide concentration (5% to 9%). The method of Zwaan was
used to estimate the molecular weight by plotting the native
protein size as a function of the ratio of percent migration
versus tracking dye in the two gels (23). The standards
used in the gel method included carbonic anhydrase, hfi.=

31,000; ovalbumin, Mr = 45,000; ovalbumin dimer, Mr =

35

90,000; bovine serum albumin (BSA) monomer, NQ.= 66,200;
BSA dimer, M = 132,000; BSA trimer, Mr = 198,600; myosin,

I."

Mr = 200,000; and B-galactosidase, Mr = 465,000.

In addition, the molecular weight for native urease was
deduced by using Superose 6 (1.0 x 30 cm, Pharmacia) gel
filtration chromatography. Standards included thyroglob-
ulin, Mr = 670,000; ferritin, Mr = 470,000; y-globulin, Mr
= 158,000; ovalbumin, Mr = 44,000; myoglobin, Mr = 17,000;
and vitamin B-12, AL.==14350. Samples were eluted at 0.5

ml min'1 in PEB buffer containing 0.1 M KCl while monitoring

the absorbance at 280 nm.

Stability studies- Urease (7 ug mail) was incubated at
O-2'C in a 10-fold dilution of assay buffer containing var-
ious concentrations of detergents, glycols, thiols, salts,
or mixtures of these substances. Aliquots from these
incubations were diluted 200-fold into the routine assay

buffer and assayed for activity.

(pH Studies- The enzyme stability was assessed at pH
values from 4 to 11 by incubating urease in buffers at
various pH values for 30 min then assaying aliquots of the
mixture in a 40-fold volume of the routine assay buffer
containing 2.5 mM to 50 mM urea. The test buffers included
acetate (pH 4-6), MES (pH 5-7), CHES (pH 8.5-10.5), or CAPS
(pH 9-12) at a concentration of 50 mM. In addition, all

buffers contained 10 mM EDTA.

36

The effects of pH on Km (urea) and Vmax were estab-
lished by assaying urease activity in buffers containing 25
mM buffer, 5 mM EDTA and 2.5 mM to 50 mM urea, at the in-
dicated pH values. The rate of urea hydrolysis was linear
at each pH value at which the enzyme was stable. Buffers
included the four mentioned above, HEPES (pH 6.5-9.0), and

phosphate (pH 5.0-11.0).

Amino acid analysis- Protein samples were hydrolyzed
in vacuo in 6 N HCl at 110°C for 24, 48, and 72 hr. Dup-
licate samples were analyzed by using a Beckman model 119CL
automatic amino acid analyzer with ninhydrin detection
system. The instrument was equipped with a model 126 data
system for peak integration at 440 and 570 nm. Cysteine was
quantitated as carboxymethyl cysteine for samples which were
denatured, reduced and alkylated by using iodoacetic acid
(2%). Threonine and serine values were extrapolated to zero
time of hydrolysis. Tryptophan was quantitated spectro-
photometrically by the absorbance at 280 nm; the extinction
coefficient was taken to be 5700 M‘hxwl for Trp and 1300

M‘lcm‘l for Tyr.

Ni analysis- The nickel content of samples was quanti-
tated by using a Perkin-Elmer PE-500 atomic absorption spec-
trophotometer equipped with an HGA 500 graphite furnace and

an AS-l autosampler. Samples were hydrolyzed in 1 M nitric

I

r-\

a:
u."

 

‘o

a \V
3AM

 

37

acid, evaporated, and resuspended in 50 mM HNO3. Nickel
standards, which in some cases contained bovine serum
albumin to provide a protein matrix, were treated iden-
tically to the urease samples. Aliquots of each sample (20
ul) were dried at 120'C, charred at 1200°C, and atomized at
2700'C. Nickel analysis was carried out using the in-
strument's background correction mode, and either peak

height or peak area were monitored.

RESULTS

Urease purification- Extracts from 190 g of cell (wet
weight) were chromatographed on a column (2.5 x 15 cm) of
DEAE-Sepharose (Pharmacia) as shown in Figure 1. A 400 ml
linear gradient from 0 M to 1 M KCl in PEB buffer was used
to recover urease as a single peak of activity at approx-
imately 0.45 M KCl. Fractions containing peak urease
activity were adjusted to 1 M KCl and chromatographed on a
column (2.5 X 15 cm) of phenyl-Sepharose (Pharmacia) by
using step elution. The column was washed with 150 ml of 1
M KCl in PEB and activity was eluted with 100 ml PEB buffer.
As illustrated in Figure 2, a broad peak of urease activity
was obtained, well separated from the major peak of protein.
The peak urease fractions were pooled, diluted with an equal

volume of PEB buffer and chromatographed on a Mono Q column

38

 

   
  

 

 

 

 

 

 

=
T «000
' 7'.
A ‘: . E -‘ 1.0
O o . I.» 1'
o ' ,u (600 s
u ‘0 : '. ' E
‘I' . I -
C . ' 9 O
. . 4 g .
5 i v -
C o 1 ‘00 I
z ‘ . I g -
3 : I > 10 5
‘ o r" :- '=‘
'3'” g u U
..... ' u 4 5!.
2° ‘ J 200 1
O. ‘
6.. ‘
..... .
Ln" 4 , —‘ 0°
360 400 500 500 7oo .00

VOLUME (ml)

Figure 1. DBAB-Sepharoae chromatography of x.
aerogenes urease. Cell extracts were chromatographed on a
column of DEAE-Sepharose by using a linear KCl gradient as
indicated ( ' '° ° ). Fractions (5.5 ml) were monitored for
absorbance at 280 nm after diluting aliquots 20-fold
(O————O), and samples were assayed for urease activity (O-—

- »O). The pooled fractions are indicated by the bar.

39

 

|.O M KCI PEB
IOOO
O
O 800
N
8 600
C
O
a
b
o
3 400
4
200

 

 

 

 

 

VOLUME ( ml )

Figure 2. Phanyl-Sepharosa chromatography of K..
aerogenes urease. The pooled fractions from Figure 1 were
adjusted to 1 M KCl and chromatographed on a column of
phenyl-Sepharose by using step elution, decreasing the
concentration of KCl as indicated while monitoring the
absorbance at 280 nm (0 ———-O). Samples (3 ml) were
assayed for urease activity (0 - - - O) and the pooled

fractions are indicated by the bar.

ACTIVITY (umol min" ml")

NH).

CEF-h<~

40

(1.0 X 10 cm, Pharmacia) As shown in Figure 3, a KCl
gradient was used to resolve urease from several contam-
inants. The Mono Q fractions containing the highest urease
activity were concentrated to 0.2 ml by using a Centricon-lO
(Amicon) and chromatographed on a column (1.0 X 30 cm) of
Superose 6 (Pharmacia) as illustrated in Figure 4. This
final purification step served to desalt the enzyme and to
provide an estimate of the native molecular weight (vide
infra). The purification of K. aerogenes urease is

summarized in Table 1.

Table 1

Urease purification from K. aerogenes

 

'Rmfl Emnmm

Pun’fication step Specific Activity Purification Total Activity Protein Recovery
(mmole urea min'1mg") (told) (mmol min") (mg) (96)

 

Cell extracts 2.06 1 19500 9460 100
DEAE-Sepharose 81.1 39 11800 145 60.5
PhenyI-Sepharose 574 279 5420 9.45 27.8
Mono Q 1170 568 5080 4 34 26.1

Superose 6 2200 1070 4770 217 24.5

 

41

 

 

8“.
.‘b
-
U
”h I
.
7 n
“In '
.D’ '
'
“I 9
“b - .I.
.
3
08)- a
-
p
”1- 2 «.80.
.0)-

 

 

 

 

 

 

 

Figure 3. none Q chromatography of.x. aerogenes
urease. The pooled fraction from Figure 2 was subjected to
Mono Q FPLC by using a KCl gradient as indicated (' °' ).
The effluent was monitored for absorbance at 280 nm (————7
and the fractions were assayed for urease activity (O - - -

O). The pooled fractions are indicated by the bar.

42

 

 

 

 

 

 

C22]
" f
.. 1.54 I: .6000 g
o 00
m '1 r
N - E
1v j E
3 1.0L 24000 g
3 i: a
0 °- 1.
m 32
3, 1' :
‘3 0.5L : f -2000 3.”
3 °. .5.
I - h
o ' U
j L‘A (
E=§= 1 . L . . it

 

 

0|
611
G

20

N
u
u
0

VOLUME (m0

Figure 4. Superose 6 chromatography of x; aerogenes
urease. The urease containing pool from Figure 3 was con-
centrated to 0.2 ml and chromatographed on a column of Su-
perose 6 resin. The effluent was monitored for absorbance

at 280 nm (———4 and the fractions were assayed for activity

(O---O).

43

SDS-polyacrylamide gel electrophoresis of K. aerogenes
urease- K. aerogenes urease was found to contain three
distinct polypeptides (Afl.= 72,000 1 2,000; 11,000 r 2,000;
and 9,000 r 2,000) by using SDS-polyacrylamide gel electro-
phoresis (Figure 5). Gels comprised of less than 10% acryl-
amide did not resolve the smaller components from the bromo—
phenol blue dye front; thus 12% or 20% acrylamide gels were
used to estimate the molecular weight of the small compon-
ents. The integrated intensities of the three bands from
Coomassie blue stained gels were used to compare the compon-
ent ratios. This method is suspect because individual
polypeptides may have different affinities for the dye;
nevertheless, this procedure yielded ratios for the
large:small:smallest bands of 1:2:2. None of the components
were glycopeptides, as demonstrated by a failure to stain

using dansyl hydrazine, a carbohydrate specific stain.

Native molecular weight for K. aerogenes urease- By
using the method of Zwaan (23) to analyze mobilities on non-
denaturing gels, the native molecular weight for K.
aerogenes urease was estimated as 190,000 r 20,000 daltons.
This value was not significantly altered if the enzyme was
incubated at room temperature in 1.0 M KCl, 5% SDS, 5%
Triton X-100, 50% ethylene glycol, 50% glycerol, 50%
sucrose, 5% B—mercaptoethanol, 20% dimethylsulfoxide, or
mixtures of these substances prior to electrophoresis.

Superose 6 gel filtration chromatography was used as a

44

ZCHD III!

 

 

 

||6 . ~ F
92.5 ‘-
45 "" I...
31 --s -' ‘3
21.5 -
14.4 -- “1...... ______,_
kD Std Kn. Rm. Su Sr

Figure 5. Sodium dodecyl sulfate-polyacrylamide gel

electrophoresis of several microbial ureases. Standard

proteins (2 #9) and ureases from K. aerogenes (K. a.), P.
mirabilis (P. m.), S. ureae (S. u.), and S. ruminantium (S.
r.) (8 pg each) were denatured and electrophoresed on a 5-

15% gradient gel, and then stained with Coomassie Brilliant
Blue. Details of these methods are described under

"Experimental Procedures".

45

second method to estimate the native molecular weight for
urease. Assuming that the ratio of molecular weight to
Stoke's radius is similar for the enzyme and the standards,
the K. aerogenes urease molecular weight is 260,000 r 50,000

daltons.

Attempts to resolve the three urease components- The
presence of three polypeptide components in K. aerogenes
urease is unprecedented when compared with previously pub-
lished urease preparations. To examine this discrepancy,
intensive efforts were carried out to further resolve the
enzyme by electrophoretic and chromatographic techniques.
Two-dimensional gel electrophoresis, in which the urease
band from native gel electrophoresis was subsequently de-
natured and run on an SDS polyacrylamide gel, was used to
demonstrate that all three polypeptides remain associated
during native gel electrophoresis. Furthermore, prior in-
cubation of urease in the presence of 1 M salt, 5% SDS, 5%
Triton X-100, 50% ethylene glycol, 50% glycerol, 50%
sucrose, 5% B-mercaptoethanol, 20% dimethylsulfoxide, or
mixtures of these substances had no apparent effect on
component association as demonstrated by native gel electro-
phoresis. Ureolytic fractions obtained by using Superose 6
chromatography equilibrated with PEB buffers containing 0.5
M KCl, 0.1% SDS, or 3% B-mercaptoethanol consistently pos-
sessed all three components upon SDS-polyacrylamide gel

electrophoretic analysis. Mono Q chromatography of active

46

urease under a variety of buffer conditions (pH values from
6.0 to 9.0, the use of 20 mM B-mercaptoethanol, and elution
with 0.5% triton X-100 in the buffers) also was incapable of
resolving the three urease components. Thus, the three
polypeptides remained associated under all conditions at
which the enzyme activity was stable. Resolution of the
large polypeptide has only been achieved by Superose 6 gel
filtration chromatography in the presence of 6 M guanidine-
HCl; however, enzyme activity was lost under these condi-

tions.

Enzyme stability studies- Attempts to resolve the
three urease components included the use of rather harsh
conditions; yet, the enzyme retained activity in most of
these experiments. The stability of K. aerogenes urease was
quantitatively examined under varied buffer conditions at 0-
2°C, as summarized in Table 2. The enzyme was found to
retain nearly full activity for over 200 h despite the
presence of salts, glycols, thiols, and detergents.

Glycols, and perhaps 1 M KCl, enhanced the ureolytic
activity at initial times; however 20-30% of the initial
values were lost after 200 h, resulting in little net change
from the control. The enzyme activity appeared to be more
stable in Triton X-100 than in SDS; nevertheless, the enzyme
could tolerate high concentrations of both detergents. In-
cubations containing B-mercaptoethanol exhibited an initial

drop in activity, consistent with competitive inhibition by

47

Table 2

Stability of K. aerogenes urease

 

 

Buffer additionsz Percent activity
remaining1
0 h 24 h 200 h

None 100 106 111
1 M KCl 119 100 97
1 M KCl, 20% glycerol 131 107 91
20% glycerol 133 115 104
20% ethylene glycol 131 119 106
200 mM B-ME3 92 104 97
l % SDS 95 118 61
1 % SDS, 200 mM B—ME 89 99 95
5 % Triton X-100 85 112 102
5 % Triton X-100, 76 103 94

200 mM B-ME

 

1 Values (accurate to i 10% are relative to the control at
zero time. Incubations were at 0-2'C.

2 Control incubation buffer was comprised of 5 mM HEPES, 0.1
mM EDTA, pH 7.75.

3 fl-ME, B-mercaptoethanol.

48

this compound (see below). Thiol oxidation may lead to the

increase in activity for these samples after 24 h.

Kinetic parameters- The rate of urea hydrolysis was
monitored as a function of urea concentration from 1 mM to
50 mM urea. Simple Michaelis-Menten kinetics were followed;
however, substrate inhibition was observed if the urea con-
centration was increased to 100 mM. The data was analyzed
by the method of Wilkinson (25) to yield a Kh for urea of
2.8 t 0.6 mM and a.igax of 2800 i 200 umol urea degraded
min'l mg'l at 37‘c in 25 mM HEPES, 5.0 mM EDTA buffer, pH
7.75. Some enzyme samples contained B-mercaptoethanol,
which was a competitive inhibitor of K. aerogenes urease;
however, the concentration of this compound in the assay

conditions led to negligible inhibition.

pH Studies- K. aerogenes urease retained full activity
when assayed under standard conditions following a 30 min
incubation at 37°C in buffers ranging from pH 5 to 10.5;
however, the enzyme was rapidly inactivated outside of this
range.

The effect of pH on the kinetic constants for urease
was determined in buffers at various pH values. The Kh for
urea was nearly constant over the pH range at which the
enzyme is stable; i.e., the Kh varied slightly from ca. 2
to 4 mM. In contrast, the “muzexhibited an optimum from pH

7.25 to 8.25. The effect of pH on log igax/Kh is shown in

49

Figure 6. The slope is 1 below the optimum and -1 above the
optimum with inflection points at pH values of 6.55 t 0.10

and 8.85 t 0.10 (26).

Amino acid analysis- The amino acid composition of K.
aerogenes urease is shown in Table 3. Also shown for com-
parison are the compositions reported for ureases from B.
ammoniagenes (9), jack bean (5), and soy bean (8). The
relatedness of K. aerogenes urease to these other proteins
was examined by using the method of Cornish-Bowden (27) in
which the difference index (DI), the compositional diver—
gence (D) and the Marchalonis and Weltman index (SAQ) were

calculated.

Atomic absorption analysis- Purified K. aerogenes
urease was demonstrated to contain nickel by atomic ab-
sorption analysis. Three independent preparations of urease
were found to contain 2.1 i 0.3 mol Ni/mol 72,000 dalton
subunit when compared to a standard curve which was prepared

as described in Methods.

DISCUSSION

The studies reported here extend the work of Friedrich
and Magasanik with K. aerogenes urease (13) and of Kamel and

Hamed with urease from the related microbe, A. aerogenes

50

 

 

 

 

i- .-
E
)6
\ '2' a-
X
0
E _ ..
:>
a
2 -3)- -
I- «Ii
4.

 

Figure 6. The effect of pH on log V.“ / K., for K.
aerogenes urease. Urease‘Kmu1and K5 values were calculated
and the log of the ratio is plotted here in arbitrary units
versus the pH of the assay buffer. The buffers include
acetate (0), MES (V), HEPES (X), CHES (A), CAPS (o), and

phosphate (+).

51

Table 3
Amino Acid Compositions of Ureases

 

Amino Acid K3 aerogenesaB. ammoniagenesbJack bean? Soybeand

 

Cys 1.44 0.38 1.79 0.47
Asx 9.23 11.60 10.60 12.76
Thr 6.90 7.23 6.55 5.28
Ser 6.05 3.64 5.48 5.79
Glx 10.32 10.22 8.22 10.09
Pro 5.50 4.66 5.00 6.09
Gly 12.35 9.68 9.40 10.35
Ala 10.45 10.68 8.80 7.44
Val 7.46 7.80 6.55 6.34
Met 1.68 1.75 2.50 2.04
Ile 5.72 7.13 7.86 6.16
Leu 6.92 7.46 8.21 8.17
Leu 6.92 7.46 8.21 8.17
Tyr 2 1.17 2.5 2.43
Phe 2.57 3.02 2.86 4.23
His 3.00 2.91 2.98 2.09
Lys 3.44 4.87 5.71 5.64
Arg 3.96 4.89 4.52 4.62
Trp 0.99 0.90 0.48 NRe
Related
in i f
D1 0 8.1 9.58 10.62
D 0 0.052 0.057 0.065
SAQ 0 26.8 32.33 41.73

 

meOU‘m

Values represent the mole percent of each amino acid.
Taken from (9).

Taken from (5).

Taken from (8).

NR, not reported.

Relatedness indices calculated for each urease compared
to the K. aerogenes urease using the method of Cornish-
Bowden (27).

52

PRL-R3 (14). K. aerogenes urease was purified 1070-fold to
homogeneity by standard chromatographic techniques, similar
to the procedure used to isolate S. ruminantium urease (11).
The final specific activity achieved for the K. aerogenes
enzyme (2200 umol ndnfl mg‘l) is among the highest reported
for any bacterial urease and approaches that of the jack
bean enzyme, about 3500 umol ndnfl mg‘1 or 93 katal/liter
[Aum (where a katal is the amount of enzyme which degrades
1 mol of urea/s in a defined pH-stat assay) (1). For com-
parison, 24-fold-purified enzyme obtained by Friedrich and
Magasanik had an activity of 45.4 umol min'lnrrl (13), and
lSO-fold-purified A. aerogenes urease had an activity of 690
umol min‘lrmr4 (14). The Kh for urea of 2.8 mM determined
for purified K. aerogenes urease was 4 times that determined
by Friedrich and Magasanik of 0.7 mM (13); however, dif-
ferent strains were employed. Similarly, the K5 was 2-fold
greater with this enzyme compared to the 1.48 mM Km deter-
mined for A. aerogenes urease (14). The Vmax calculated for
K. aerogenes urease was 2800 umol min‘l mg’l.

A novel feature of K. aerogenes urease is the presence
of three polypeptide components in the active enzyme. One
of these species (hfi.= 72,000) is similar in size to many
other microbial ureases, whereas the two smaller components
(45.: 11,000 and 9,000) have not been described in any
other urease enzyme. Intensive efforts were used to resolve
the larger polypeptide from the smaller components; however,

the three species remained associated under all conditions

53

which retained urease activity (presence of salt, glycols,
thiols, detergents, and mixtures of these substances).

These results may indicate that all three polypeptides are
subunits of K. aerogenes urease. The three polypeptides are
unlikely to be proteolytically derived from a common pre-
cursor found in cell extracts, because a variety of protease
inhibitors (toluenesulfonyl flouride, bestatin, pepstatin,
leupeptin, aprotinin, phosphoramidone, and EDTA) had no ap-
parent effect on the SDS-PAGE gel profile for the purified
enzyme. The two smaller polypeptides run at the dye front
in gels containing less than 10% acrylamide, and these
species can be easily overlooked. Indeed, we have found
that the S. ruminantium urease, previously thought to
possess a singly 70,000-da1ton polypeptide (11), also ex-
hibits two small polypeptides on 20% acrylamide gels.
Furthermore, in our most highly purified samples of urease
from Proteus mirabilis (provided by Julie M. Breitenbach)
and Sporosarcina urea (provided by Rick Ye) we have observed
a large subunit and two small polypeptides (Figure 5).

Thus, one large and two small polypeptides may be generally
associated with microbial ureases. Recent studies involving
the cloned urease gene from Providencia stuartii support
this hypothesis (28). Mobley et al. (28) found that there
may be at least two polypeptides peptides (M1. = 73,000 and
25,500) associated with urease in this microbe by using
transposon mutagenesis. In contrast to the above microbial

enzymes, we have not observed any small polypeptides

54

associated with jack bean urease. Further efforts are
clearly required to establish the roles for each of the
three K. aerogenes urease components.

The quaternary structure and native molecular weight of
K. aerogenes urease were not precisely established. From
the integrated intensities of the gel scan profiles, an ap-
proximate stochiometry for the three components was found to
be 1:2:2. This result, combined with the native molecular
weight determined by gel electrophoresis and Superose 6 gel
filtration chromatography, would indicate a native urease
structure (Afi.= 224,000) containing 2 mol of the 72,000-
dalton peptide, 4 mol of the 11,000-dalton peptide and 4 mol
of the 9,000-dalton component. These results contrast sig-
nificantly with the simple homohexameric structure of jack
bean urease and the suggested homopolymeric structures
(trimer, tetramer, pentamer, and hexamer) reported for other
microbial ureases (9, 10, 11, 29, 30).

Stability studies of ureases from both jack bean (1)
and Arthrobacter oxydans (31) showed irreversible loss of
activity below pH 4.5. Similar results were observed with
the K. aerogenes enzyme; in addition, a high pH inactivation
occurred at pH 10.5 and above. As in the case of the jack
bean urease (32), the Kh for urea is nearly constant
throughout the entire pH range over which the enzyme is
stable. K3 aerogenes urease exhibited a VhaxfiKh dependence
on pH which is most easily interpreted by assuming a

catalytic requirement for a deprotonated active site group

55

with a pK' = 6.55 and a protonated group with a pr = 8.85.
Very early work with jack bean urease indicated similar
simple behavior with pk; values of 6.1 and 9.2 (33); how-
ever, subsequent studies suggest a more complex behavior
(32). In contrast to K. aerogenes urease, the urease from
A. aerogenes exhibited very complex pH behavior and was
suggested to possess three functional groups with pk; values
of 5.6, 6.65, and 8.1 (14).

The amino acid composition for K. aerogenes urease was
determined and compared to that of jack bean, soy bean, and
B. ammoniagenes ureases. K. aerogenes urease was not re-
lated to the plant enzymes according to the "weak" test of
Cornish-Bowden (27); i.e. the relatedness indices exceeded
the recommended limits for a sequence of 840 amino acids.
Surprisingly, the two microbial urease compositions also
fail to meet the weak test for relatedness; however, the D1,
D, and SAQ values indicated that K. aerogenes urease was
more closely related to the bacterial than the plant
enzymes.

K. aerogenes urease has joined the growing list of
nickel-containing enzymes which include other ureases,
methylcoenzyme M reductase, carbon monoxide dehydrogenase,
and certain hydrogenases (35). Thus far, nickel has been
shown to be present in ureases from jack bean (6), say bean
(8), S. ruminantium (11), B. ammoniagenes (9), B. pasteurii
(10), and A. axydans (30). Furthermore, indirect evidence

of nickel-dependent growth is consistent with the presence

56

of nickel in ureases from other plants (34) and from S.
ureae (31), Aspergillus nidulans (36), Phaeadactylum tri-
cornutum (37) and Tetraselmis subcordifarmis (37). The
nickel stochiometry is clearly 2 mol nickel/mol subunit in
the jack bean and soybean enzymes, whereas either 1 mol
(9,10) or 2 mol of nickel (11) per subunit has been found
for the bacterial ureases. The K. aerogenes urease results
are most consistent with 2 mol nickel/mol of the 72,000-
dalton subunit. The environment of nickel and its role in

urease hydrolysis have not been determined.

10.

ll.

12.

13.

14.

References

Andrews, R. R., Blakeley R. L., & Zerner, B. (1984) in
Advances in Inorganic Biohemistry, (Eichhorn, B. L.,
and Marzilli, L. G., eds) Vol. 6, pp.245-283, Elsevier
Scientific Publishing Co., New York

Blakeley, R. L., & Zerner, B. (1984) J3 Mal. Catal.
23. 263-292

Sumner, J. B. (1926) J. Biol. Chem. 69, 436-441
Dixon, N. E., Hinds, J. A., Fihelly, A. K., Gaola, C.,
Winzor, D. J., Blakeley, R. L., & Zerner, B. (1980)

Can. J. Biochem. 58, 1323-1334

Mamiya, G., Takishima, K., Masakuni, M., Kayumi, T.,
Ogawa, K., & Sikita, T. (1985) Proc. Jpn. Acad. 61,
395-398

Dixon, N. E., Gazzola, C., Blakeley, R. L., & Zerner, B.
(1975) J; Am. Chem. Soc. 87. 4131-4133

Alagna, L., Hasnain, S. S., Piggott, B., & Williams, D.
J. (1984) Biochem J. 220. 591-595

Polacco, J. C., & Havir, E. A. (1979) J. Biol. Chem.
254, 1707-1715

Nakano, H., Takenishi, S., & Watanabe, Y. (1984)
Agric. Biol. Chem. 48, 1495-1502

Christians, S., & Kaltwasser, H. (1986) Arch.
.Micrabiol. 145, 51-55

Hausinger, R. P. (1986) J. Biol. Chem. 261, 7866-
7870

Rosenstein, I. J., Hamilton-Miller, J. M., & Brumfitt,
w. (1981) Infect. Immun. 32, 32-37

Friedrich, B., & Magasanik, B. (1977) J. Bacterial.
131, 446-452

Kamel, M. Y., & Hamed, R. R. (1975) Acta Biol. Med.
Ger. 34, 971-979

57

 

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29

30.

31.

32.

33.

58
Smith, C. J., Hespell, R. B., & Bryant, M. P. (1980)
J. Bacteriol. 141, 593-602
Weatherburn, M. W. (1967) Anal. Chem. 39, 971-974

Lowry O. H., Rosebrough, N. J., Farr, A. L., &
Randall, R. J. (1951) J. Biol. Chem. 193. 265-275

Lammeli, U. K. (1970) Nature 227, 680-685

Guilian, G. G., Moss, R. L., & Greaser, M. (1983)
Anal. Biochem. 129, 277-287

Eckhardt, A. E., Hayes, C. E., & Goldstein, I. J.
(1976) Anal. Biochem. 73. 192-197

Blattler, D. P., Contaxis, C. C., & Reithel, F. J.
(1967) Nature 216, 274-275

Fishbein, W. N. (1969) 5V9 International Symposium
Chrom. Elect., pp. 238-241, Ann Arbor-Humphrey Science
Press, Ann Arbor, MI

Zwaan, J. (1967) Anal. Biochem. 21, 155—168

Gurd, F. R. N. (1967) .Methads Ehzymal. 11, 532-541
Wilkinson, G. N. (1961) Biochem. J. 80, 324-332

Siegel, I. H. (1975) Enzyme Kinetics, John Wiley &
Sons, New York

Cornish-Bowden, A. (1983) .Methads Ehzymol. 91, 60-75

Mobley, H. L. T., Jones, B. D., & Jerse, A. E. (1986)
Infect. Immun. 54, 161-169

Carvajal, N. C., Fernandez, M., Rodriguez, F. P., &

Donoso, M. (1982) Phytachemistry (Oxf.) 21, 2821-
2823

Creaser, E. H., & Porter, R. L. (1985) Int. J.
Biochem. 17. 1339-1341

Schneider, J., & Kaltwasser, H. (1984) Arch.
.Micrabiol. 139, 355-360

Dixon, N. E., Riddles, P. W., Gazzola, C., Blakeley, R.
L., & Zerner, B. (1980) Can. J. Biochem. 58, 1335-
1344

Laidler, K. J. (1955) Faraday Soc. Trans. 51, 550-
561

34.
35.
36.

37.

59

(1977) Plant Sci. Lett.

Polacco, J. C.

Hausinger,

MacKay, E. M., & Pateman,
Microbial. 116, 249-21
Rees, T. A. B., & Bekheet,
(Berl.) 156. 385-387

J. A.

I.

A.

R. P. (1987) .Micrabiol. Rev.

(1980)

(1982)

10, 249-255

51. 22-42
J. Gen.

Planta

Competitive Inhibitors of Klebsiella aerogenes Urease:

Mechanism of Interaction with the Nickel Active Site

ABSTRACT

We examined several compounds for their mechanisms of
inhibition with the nickel-containing active site of homo-
geneous Klebsiella aerogenes urease. Thiolate anions compe-
titively inhibit urease and directly interact with the met-
allocenter, as shown by UV-visible absorbance spectroscopic
studies. Cysteamine, which possesses a cationic B-amino
group, exhibited a high affinity for urease (K,- = 5 )LM) .
whereas thiolates containing anionic carboxyl groups were
uniformly poor inhibitors. Phosphate monoanion competi-
tively inhibits a protonated form of urease with a pk; of
less than 5. Both the thiolate and phosphate inhibition re-
sults are consistent with charge repulsion by an anionic
group in the urease active site. Acetohydroxamic acid (AHA)
was shown to be a slow-binding competitive inhibitor of
urease. This compound forms an initial E-AHA complex which
then undergoes a slow transformation to yield an E-AHA*
complex; the overall dissociation constant of AHA is 2.6 pM.
Phenylphosphorodiamidate, also shown to be a slow-binding
competitive inhibitor, possesses an overall dissociation

constant of 94 pM. The tight binding of

60

nae

vVl

.aé‘

'“l

-4.1

up:

AA.
vv.

Q...

 

61

phenylphosphorodiamidate was exploited to demonstrate the
presence of two active sites per enzyme molecule. Urease
contains 4 mol nickel/mol enzyme, hence there are two mol
nickel ion/catalytic unit. Each of the two slow-binding
inhibitors are proposed to form complexes in which the
inhibitor bridges the two active site nickel ions. The
inhibition results obtained for K. aerogenes urease were
compared with inhibition studies of other ureases and are
interpreted in terms of a model for catalysis proposed for
the jack bean enzyme (Dixon, N. E., Riddles, P. W., Gazzola,
C., Blakeley, R. L., and Zerner, B. (1980) Can. J. Biochem.

58, 1335-1344.)

INTRODUCTION

Urease (urea amidohydrolase E. C. 3.5.1.5) is a nickel-
containing enzyme that hydrolyzes urea to form carbamate and
ammonia; carbamate spontaneously degrades to C02 and a sec-
ond molecule of ammonia (1-3). The best characterized
urease is the enzyme purified from the plant Canavalia
ensifarmis (jack bean) (1,2). Jack bean urease is a homo-
hexameric enzyme (subunit An = 90,770; 4) which contains 2
mol nickel/mol subunit (5). Zerner and colleagues have pro-
posed a model for jack bean urease catalysis in which one
nickel coordinates the oxygen atom of urea, polarizing the
carbonyl group, and a second nickel coordinates hydroxide
ion, the catalytic nucleophile. Three amino acid residues
are also proposed to be located at the active site (6): a
carboxyl group, a sulfhydryl group, and an unidentified
base. Despite extensive investigation the detailed mecha-
nism of catalysis has not been established and the structure
of the metallocenter is unknown.

In contrast to the wealth of studies on the jack bean
enzyme, much less in known about catalysis by microbial
ureases (3). Clear structural differences from the plant
enzyme have been observed; for example, the bacterial urease
of Klebsiella aerogenes possesses three subunit types (AA.

= 72,000, 11,000, and 9,000) in an 012B474 stoichiometry (7)

62

63

and contains 4 mol nickel/mol enzyme (7). Further charac-
terization of microbial urease is important because infec-
tious ureolytic microorganisms can contribute to the devel-
opment of urinary stones, pyelonephritis, gastric ulcera-
tion, catheter encrustation, and other pathogenic conditions
(3). In addition, elevated levels of soil microbial urease
activity can decrease the efficiency of urea fertilizers (8)
and retard seed germination and seedling growth (9). Thus,
the study of urease inhibitors may have medical or agronomic
significance, as well as providing insight into the urease
catalytic mechanism.

Here we characterize the inhibition of purified K.
aerogenes urease by thiols, phosphate, acetohydroxamate and
phenylphosphorodiamidate. We compare these results to pre—
vious bacterial urease inhibition studies, and we discuss
our results in terms of the catalytic model proposed for

jack bean urease.

EXPERIMENTAL PROCEDURES

Materials- Sources are: acetohydroxamic acid, bovine
serum albumin, cysteamine, cysteine methyl ester, B-mercap—
toethanol (Sigma); phenylphosphorodiamidate (ICN Biomedicals
Inc.); urea (enzyme grade) (U. S. Biochemical Corp.); ethane
thiol, 3-mercapto-2-butanol, 2-mercaptopropionate, 3-mercap-

tapropionate, methylthioglycolate, 2-propane thiol,

64

3-mercapto-1,2-propanediol (Aldrich); cysteine (Fisher);

cresol red (Allied Chemical and Dye Corp.).

Enzyme purification- Urease was purified from K.
aerogenes (currently K; pneumoniae) strain CG 253 carrying
the recombinant plasmid pKAU19 (10) by procedures essen-
tially identical to those described previously (7). Since
the initial specific activity was much higher in this over-
producing organism, the final gel filtration step was not
required for maximal specific activity (2500 umol min-1
mg'l). The.Kh'value for urease was reevaluated to be 2.4 mM
at pH 7-9 rather than 2.8 mM as reported initially (7). The
Kh decreases moderately to 0.72 mM at pH 5-6. The enzyme
appeared homogenous by sodium dodecyl sulfate polyacrylamide
gel electrophoresis. Enzyme concentrations were based on a

molecular mass of 224 kDa (7).

Assays- Unless otherwise noted, urease was assayed in
the presence of 5 mM EDTA, 50 mM HEPES, pH 7.75 and 50 mM
urea at 37'C (standard assay mixture). Reactions were in-
itiated by addition of small aliquots ( g 0.1 ml) of enzyme
to standard assay mixtures ( 2 2.0 ml) at 37°C. Portions of
the assay mixtures were withdrawn at intervals and the
ammonia was converted to indophenol (11) which was quanti-
tated by measuring the absorbance at 625 nm. Linear regres-
sion analysis of the ammonia concentrations versus time gave

the initial rate.

65

Alternative assays were used to measure urea hydrolysis
when compounds interfered with the indophenol ammonia assay.
Cysteine inhibition was examined by using an ammonia micro-
electrode (Microelectrodes Inc.). Standard assays with
varying urea and cysteine concentrations were run, and the
reaction was terminated by addition of 0.1 ml of 2.0 M HNO3
to 0.8 ml aliquots of these assays. A low pH was chosen to
terminate the reaction because urease is irreversibly inact-
ivated below pH 4.5 and because the ammonia exists as the
nonvolatile ammonium ion at low pH values. Ammonia concen-
tration was determined immediately after aliquots were made
1.0 M in KOH. Millivolt readings were converted to concen-
tration by comparison to an NH4C1 standard.

A Spectrophotometric assay was employed to examine the
effect of thiourea on urease activity. Urease was added to
a 1.0 ml cuvette at 37'C containing 10 mM Tris-HCl pH 7.7,
80 ug/ml cresol red, and varied concentrations of urea and
thiourea. Urea hydrolysis was monitored by the increase in
the absorbance at 573 nm, which was linear with time between
0.8 and 1.6 absorbance units. Initial rates were calculated
by comparison to absorbance changes in a control to which
aliquots of KOH were added.

Protein was assayed by the method of Lowry (11) with

bovine serum albumin as the standard.

Spectrascapy- All spectra were obtained by using a

Gilford Response spectrophotometer with a 1.0 cm

66

microcuvette (25‘C). Spectral data were transferred to an
IBM-PC to calculate difference spectra. Enzyme samples were
extensively dialyzed prior to scanning to remove the B-ME

present during enzyme storage.

RQ.Determinations— For inhibitors which equilibrate
rapidly with urease, initial rates were determined in the
presence of four different inhibitor concentrations at each
of seven urea concentrations (50, 10, 5, 3.3, 2.5, 1.667,
and 1.25 mM). Data were plotted according to Lineweaver—
Burk (13), and best fit lines were calculated by using the
method of Wilkinson (14). Kfi‘was determined from replots of
slope or apparent Km values versus inhibitor concentration
(15). The response of Rh to changes in pH was investigated
by using 100 mM MES (pH 5.00-6.75) or 100 mM HEPES (pH 6.75-
8.50) buffers, with 5 mM EDTA, inhibitor and urea (as

indicated above).

Inhibitor Binding Rates in the Absence of Substrate-
Urease was added to 10 mM EDTA, 100 mM HEPES, pH 7.75 at
37°C containing the indicated concentrations of slow bind-
ing, competitive inhibitor. At regular intervals, small
aliquots were diluted 40-fold into the standard assay mix-
ture, and activities were determined. Linear regression
analyses of the initial rates provided correlation coef-
ficients r > 0.995, indicating that further inhibitor bind-

ing did not occur and that very little enzyme-inhibitor

67

complex dissociated during the assays. Apparent rate con-

stants were calculated by linear regression analyses of the
data plotted as ln(V,/V0) versus time (V, = initial velocity
after t min preincubation with inhibitor and V; = initial

velocity of enzyme in the absence of inhibitor).

Progress Curves in the Presence of Inhibitar- Progress
curves for the rate of urea hydrolysis in the presence of
inhibitors were obtained by measuring the ammonia produced
as a function of time (Pg. For each experiment, 25-50 time
points were taken. A curve-fitting program (see below) was
used to fit the data points to:

P,=P0+vf:+(V,-- Vf)(1-e*m‘)&;;p (1)
where P; = product present at time 0, i? = steady state
velocity, VG = initial velocity, and hwp = apparent rate

constant (16).

Dissociation Rate- The rate of dissociation for the
enzyme-inhibitor complex (which was <2% active) was measured
by diluting 10,000-fold into an assay mixture at 37'C con-
taining the indicated concentrations of urea in 10 mM EDTA,
100 mM HEPES, pH 7.5. The concentration of AHA during the
assay was less than 0.1 uM, and the concentration of PPD was

less than 0.1 nM. Progress curves for the recovery of

68

activity were monitored for several hours and fit to
Equation 1.

Dissociation of AHA-urease was also determined for the
isolated urease-AHA complex at 37°C, following removal of
excess AHA on a Bio-Gel P-6 (0.7 x 14.0 cm) gel filtration
column. Dissociation was followed by periodically diluting
a small aliquot into the standard assay mixture and com-
paring the activity to an appropriate control. The rate of
dissociation.(&) was obtained by fitting the fraction ac-

tivity recovered as a function of time (Efl to>Equation 2.

F,=F0+(I-F0)(1-e"‘) (2)

Statistical Methods--Results are provided as the mean :
S.D., where applicable. The fitting software (17) is avail-
able free of charge to those with a modem. The BASIC pro-
gram uses the Marquardt Algorithm (18) to fit by least-
squares a set of data points to a curve. The user supplies
a function and the partial derivatives with respect to each

parameter to be fit.

Quantitation of Active Sites- The graphical method of
Ackerman-Potter (19) was used to determine the number of
active sites per native enzyme molecule. Urease (0.0—10.0
nM) in 100 mM HEPES, 10 mM EDTA, 0.1 mM B-ME, pH 7.75 was

incubated with 0, 5, or 10 nM PPD at 0°C. Aliquots were

69

analyzed for activity at 37°C after 2, 24, and 96 h by using

the standard assay mixture.

RESULTS

Thiol Inhibition- Several thiol compounds were shown
to be competitive inhibitors of urease, with apparent K;
values as provided in Table 1. To determine whether the
protonated thiol or the thiolate anion was the actual in-
hibitory species, the pH dependence of cysteamine inhibition
was characterized (Figure 1). Log Kiresponded linearly to
pH changes, with a slope of -1 at pH values below the cyste-
amine pk; of 8.1 and with a slope of 0 at pH values above
this pkh. This behavior is consistent with the deprotonated
thiol acting as the inhibitory species.

To compare relative inhibition within the series of
thiols in Table 1 the pH dependence of each urease inhibitor
should be determined. However, in this initial analysis,
the K; values were simply recalculated on the basis of the
actual thiolate anion concentrations at pH 7.75 by using
published pKh'values. Of the compounds tested, the thiolate
with the lowest.K;value (5.0 uM) is cysteamine, which has a
positively charged B-amino group at pH 7.75. In contrast,
thiol compounds containing an anionic carboxyl group are
poor inhibitors at pH 7.75 (e.g. 3-mercaptopropionate and

cysteine, with K,- values of >500 uM and 9600 1114,

 

7O

 

 

 

 

 

—2 W I I I I I I I I I I I I I I I T T I T I I rI I T I I T] lj
I- -1
)- .1
. 1
..
_3_ _.
F d
A .. ..
o- v- .
X - .
V —4.— _
O) r- 4
O )- -1
_ - -1
- -1
_5_ l
h d
D d
r- -4
1- -1

_6 L 1 1 l 1 n 1 111 1 111 L+ 1+1 LLIJ_L

5.0 6.5 ‘7.0 7515.0 8.5

pH

Figure 1. pH dependence of. cysteamine inhibition of
urease. The Ki values for competitive inhibition of urease
by cysteamine were determined in MES (On!) or HEPES (l--

I) buffers and are plotted as log(K,-) versus pH.

71

Table l

Thiol Inhibition of K. aerogenes urease

 

 

 

Thiolate
Total anion
Inhibitor Ref.
Thiol K3 K! M.
W M

cysteamine 0.010 5.0 8.1 20
B—mercaptoethanol 0.55 7.7 9.6 20, 21
3-mercapto—l,2-propane diol 1.7 21 9.66 22
ethanethiol 21 37 10.5 20
cysteine methyl ester 0.12 110 6.53 20
3-mercapto-2-butanol 13.5 190
methylthioglycolate 1.6 850 7.7
2-propane thiol > 100 > 140 10.6 21
cysteine 95 9600 8.7 20
3-mercaptopropionate > 100 > 500 10.05 20, 21, 23

 

a) Based on total inhibitor concentration regardless of species present at pH
7.75.

b) Thiolate anion K,- values at pH 7 .75 are calculated based upon the literature pKa

values.

72

respectively). Esterification of the carboxyl group en-
hances binding as demonstrated by cysteine methyl ester with
a.K;cfi3110 uM. Thiols with uncharged side chains possess
intermediate Kivalues ranging from 7.7 uM to 190 nM. The
ability of 3-mercapto-2-butanol to inhibit urease indicates

that the active site must be accessible to secondary thiols.

Spectral interactions of B-ME with urease- In the
absence of thiols, K. aerogenes urease displays an absorp-
tion spectrum (Figure 2) which includes a peak at 406 nm (e
= 2,240 M‘lcm‘l, based on Mr = 224,000). This spectrum is
essentially pH-independent (data not shown). Binding of
thiolate anions enhances the absorbance in four regions
between 800 and 300 nm with the exact wavelength depending
upon the thiolate structure. For example, the difference
spectra maxima were 322, 374, 432, and 750 nm for B-ME,
versus 339, 387, 425, and 750 nm for cysteamine. These
spectral interactions are similar to those seen with known
thiolate-nickel complexes (e.g. nickel-substituted metal-
loenzymes or model compounds; 24).

A double reciprocal plot (Figure 3) of the absorbance
changes versus B-ME concentration provided the binding con-
stant (Ky = 0.38 i 0.12 mM) and the extinction coefficients
for the thiolate-nickel complex.(8322,ml= 2230 i 220 M'1
cm-l, 6374 mm = 1060 M-1 cm-l, 8432 m = 530 :t 80 M-1 cm-l).
The slight upward curvature at low B-ME concentrations in

Figure 3 is a result of the high protein concentration (0.25

73

 

Absorbonce

 

 

 

 

A Absorbonce

 

 

 

l—rTI'UUIUIl‘ VI'TY TIIII'I'I'

300 400 500 600 700 800
Wavelength (nm)

Figure 2. UV-visible spectra of urease in the presence
of thiol. A, Spectra of urease (56 mg/ml) in 10 mM EDTA,
100 mM HEPES, pH 7.75, at 25°C are shown in the presence of
0.0 mM( ), 0.33 mM (- - . - ), and 5.0 mM (----) B-ME.
B, Difference spectra of urease in the presence of 0.2,
0.25, 0.33, 0.5, 1, or 5 mM B-ME were calculated versus the
control (0.0 mM B-ME, 56 mg/ml).

 

74

2513- I

 

(A Absorbonce)"

 

 

TTTV ii

T ..T....,....,...fr....,.....
-2£5-—1.5 -{15 GAS 1.5 215 :3f5 415 5L5

] (ml/1")

 

[B—ME

Figure 3. Double-reciprocal plot of llA absorbance
versus 1/[B-MB]. The inverse change in absorbance at 25°C

was plotted as a function of 1/[B-ME] at A432 nm (0) . A374 m

(I). and 1432211,“ (A).

75

mM) required to observe these absorbance changes. A Dixon
plot (15), used when the enzyme and inhibitor concentrations
are of the same magnitude, yielded the same Kg value (data
not shown). The spectrally determined Ky value is identical
to the K,- determined kinetically for B-ME inhibition of urea

hydrolysis at pH 7.75 (0.55 i 0.08 mM).

Phosphate Inhibition- Lineweaver-Burk plots demon-
strated that phosphate competitively inhibits urease at pH
values from 5.0 to 7.0 (data not shown). Urease is labile at
pH values below 5.0 and phosphate inhibition was not exam-
ined. Above pH 7.0 phosphate inhibition was not purely
competitive, possibly due to the effects of high ionic
strength. Within the pH range examined, values of log K}
are pH dependent (Figure 4), exhibiting a slope of 1 from pH
5.0 to 6.3 and a slope of 2 from pH 6.3 to 7. Thus, phos-
phate inhibition of urease appears to require protonation of

two ionizable groups with pk; values of 6.3 and <5.

Acetohydroxamate inhibition- Progress curves for urea
hydrolysis in the presence of AHA are shown in Figure 5.
These curves demonstrate a gradual decrease in urease activ-
ity in the presence of AHA. The concentration of both AHA
and urea appear to affect the rate of inhibitor binding.
These initial results are consistent with slow-binding com-
petitive inhibition. Such inhibitors are thought to follow

the mechanism shown below (16), where inhibitor (I) competes

76

 

llJllllllllLl4l

'09 (K1)

4 A A 1 l

 

 

 

 

Figure 4. pH dependence of phosphate inhibition. The

K,- for competitive inhibition of urease by total phosphate

species was determined in MES (O) or HEPES (D) buffers and

is shown as log (Ki) versus pH.

 

mM NH4

 

 

1.21 :32
. .
O l-
.” :28
1.0- -
. .
. - ;2.4
E . 2
0.8. . '.: E
8 I "2.0 II. N
H ' . : 2: H
I .
1-n 0.51 ‘-1.6 E 1—0-1
53 ‘ A a a)
L A A S
'3' 0.4- :1'2 “'
;0.6
4" .. .
0.2i {I 'v' .04
. h .
4"€—:--:’---:-1 F
0.0 :11....1.... . .1 1 1 11 1.0.0

 

Figure 5. Urease progress curves in the presence of
AHA. Urease was assayed at 37°C in 10 mM EDTA, 100 mM
HEPES, pH 7.75 containing 20 mM urea (solid symbols) or 2 mM
urea (open symbols), and 0.0 mM (circles), 1.5 mM (squares).
or 3.0 mM (triangles) AHA. The reactions were initiated by
addition of urease (with 20 mM urea, [urease] = 140 pM; with
2 mM urea, [urease] = 280 pM), and the product, ammonia, was
monitored. The curves were fit as described under

“Experimental Procedures."

78

with substrate (S) for enzyme (E) to form an initial
enzyme-inhibitor complex (E-I), which is slowly transformed

to a more stable complex (E-I*).

. It,

/.

a
\ ElizEl’ Scheme I

A series of experiments was designed to test this mechanism
and to define the rate constants.

The effect of AHA concentration on the apparent rate of
inhibitor binding in the absence of substrate is shown in
Figure 6. At each inhibitor concentration, a first order
loss of activity was observed. When the apparent rate con-
stants derived from these and other experiments were replot-
ted as a function of AHA concentration, they yielded a hy-
perbolic curve (Figure 6, inset). Morrison and Walsh (16)
have shown that this behavior is expected for an inhibitor
which reacts as illustrated in Scheme I. The apparent rate

of inhibitor binding (4 follows Equation 3,

app )

 

 

[I]
Ki
Ki + l-+ Km
where S = substrate, 551:: “2 + LQ/il = Michaelis constant,

and K,- = 114/£3. Expanding from this equation, one can show

that if

79

 

 

O
O
C)
(D
(D
(D
C)
d.
L

  

 

"[AHA] (mM)

5 10

U T T l r T

0.0 0.4 'oiar 7,2. 1.6'

—4.o.

 

 

 

Time (minutes)

Figure 6. Rates of AHA binding in the absence of

substrate. ln(VflW§) is shown as a function of time for
urease (3.5 nM) incubated at 37°C and pH 7.75 in the pre-
sence of 0.0 mM (0), 1.0 mM (I), 2.0 mM (A), 3.0 mM (V),
5.0 mM (X), or 10.0 mM (0) AHA. Inset, apparent binding

rates are displayed as a function of AHA concentration.

8O

 

 

 

 

[I]
£6 << £5 [S] (Eq- 4)
[I] + K;(l+ if.)
then
[I]
Bapp ~ £5 [5] (Eq. 5)
[I] + K;(l +‘E—')
Hence,
S
l l Ki(l+£K—]-
~ — + "' (Eq. 6)
lapp is is [I]

and a double reciprocal plot of 1/itc.ipp versus 1/[I] should
yield a straight line. The data obtained using AHA exhibit
this behavior (Figure 7). The y intercept provides a value
fem-£5 = 0.047 r 0.002 s‘1 and the x intercept provides a
value for K;==1.4 i 0.2 mM.

Figure 7 also shows the result from a similar exper—
iment carried out in the presence of B-ME, a competitive
inhibitor of urease (see above). B-ME competes with AHA
binding; the Ki value (0.26 i 0.09 mM) is in reasonable
agreement with the K,- for competitive inhibition of urea
hydrolysis by B-ME (K;=:0.55 i 0.08 mM).

To demonstrate that AHA competes directly with urea,

rate constants for AHA binding at several urea

81

‘999 .

(sec)

 

 

YT rr fi‘

—1.o —o.5 0.6"015 136 115' '210

__J___. (""WM)
[AHA]

Figure 7. Double reciprocal plot of llltapp versus

1/[AHA]. Data from Figure 6 (inset) are replotted as
1/11app versus 1/[AHA] (O) . Analogous experiments were
carried out in the presence of 1.0 mM B-ME (X). Also shown
are apparent rate constants for AHA binding from experiments
such as Figure 5 for assays containing 5 mM (0). 10 mM (V),

15 mM (A), and 20 mM (I) urea.

82

concentrations were obtained from progress curves such as
those shown in Figure 5. The initial velocity did not
appear to be affected by the presence of inhibitor; however,
iapp responded hyperbolically to increasing AHA levels at
each urea concentration (data not shown), similar to the
response of hum seen in the absence of substrate (Figure 6,
inset). Double reciprocal plots (Figure 7) show that urea
competes with AHA binding, and this chi'1.6 i 0.2 mM is in
close agreement with the kinetically determined Kh value of
2.4 i 0.4 mM.

The rate of E-I* dissociatirnn 46, was determined by
diluting the AHA-urease complex (or an appropriate control)
into assay mixtures and measuring urea hydrolysis at 37°C as
a function of time (Figure 8). The resulting progress
curves (fit to Equation 1 with Vf 2 V0) yielded values for 15
of 9.4 i 1.3 x 10'5 s’1 in 40 mM urea and a.is<af 9.0 i 0.8 x
10‘5 s‘1 in 5 mM urea.

An independent determination of ”‘6 was achieved by sep—
arating the AHA-urease complex from excess AHA by gel fil-
tration and measuring activity as a function of time at
37°C. limais determined by using Equation 2 was 7.8 i 0.6 x
10'5 5'1, in close agreement with the result obtained in the
presence of substrate. Thus the approximation of Equation 4
is valid for the AHA and urea concentrations used to study
the binding rates. Also consistent with the proposed mech-
anism, substrate concentration did not significantly affect

16.

 

1.6- -

.- O
r
1.2- -
g. a
II: ,.'
:2: ’ --
2 ' j
E

 

 

ALLLLA

400

 

Time ( minutes )

Figure 8. Progress curves for AHA dissociation.
Urease which had been incubated with 1.0 mM AHA (open
symbols) or buffer (solid symbols) was diluted 10,000-fold
and assayed in the presence of 40 mM urea (circles) or 5 mM
urea (squares) at 37°C and pH 7.75. The final concentration
of urease was 8.1 pM. Curves were fit as described under

”Experimental Procedures."

84

The overall inhibition constant for a slow-binding com-
petitive inhibitor as expressed by Morrison and Walsh (16)
is

1- £6

K. = Ki (Eq. 7)

£6+£’5

H-

For AHA inhibition of K. aerogenes urease, the K: = 2.6
0.4 uM.

In other experiments (data not shown), the kinetic con-
stants for AHA binding were examined as a function of tem-
perature and pH. Kgrtmained constant as the temperature
varied between 37°C and 0°C, whereas (15 decreased 72 fold
(is = 6.45 i 0.36 x 10‘4 s'1 at 0°C). The values fom'is and
K,- were nearly unchanged when measured at pH 6.75, 7.75, and
8.25; however, 45 increased at lower pH values (e.g. £5 =
17.2 x 10‘5 at pH 6.50), decreasing the overall affinity of
AHA for urease (Equation 7). The resulting increase in K;

could explain the apparent decrease in AHA inhibition seen

by Hase and Kobashi (25) at low pH.

PPD Inhibition- Urease inhibition by PPD was also
found to exhibit slow-binding kinetics. PPD experiments es-
sentially identical to those described above for AHA were
carried out yielding K,- = 95 i 10 nM, 45 = 0.074 :t 0.005
s‘l). As with AHA, competition for PPD binding by urea (K;
= 1.43 i 0.15 mM) was in reasonable agreement with the ki-

netically determined Kh of urea (2.4 i 0.4 mM). The rate

85

of urease—PPD dissociation (determined by assaying the

complex in 250 mM urea) was 15:: 4.7 x 10‘55‘1, thus the

approximation of Equation 5 is valid for PPD. Using the Kb
#

isenuiifi values determined above, Equation 7 gives K} = 94 i

10 pM.

Quantitation of Active Sites- The tight binding inhi-
bition of urease by PPD was exploited to quantitate the num-
ber of active sites per native enzyme molecule by using the
graphical method of Ackerman-Potter (19). As illustrated
in Figure 9, incubation of PPD with urease led to a shift in
the position of the velocity versus enzyme concentration
curve. Extrapolation of the linear portion of the curve to
zero velocity demonstrates that 5.0 nM PPD completely inhi-
bits 2.3 nM enzyme and that 10.0 nM PPD completely inhibits
4.3 nM enzyme. Assuming that 1 mol PPD binds/mol active
site, the enzyme contains 2.2 i 0.2 catalytic sites/224 kDa.
Previous work (7) had demonstrated that each mol of enzyme
contained 4 mol of nickel, thus two nickel ions are present
per catalytic unit. This metal content per active site is

identical to that reported for jack bean urease (26).

Other Inhibitors- Boric acid and phenyl borate were
found to be simple competitive inhibitors of K. aerogenes
urease (K; = 0.33 mM and 11.5 mM, respectively). These
\Nalues are several fold greater than those reported for

Ffiroteus mirabilis urease (27). As in the P. mirabilis

86

Inhibition of Urease by PPdA

Velocity

umol min"ml"

O
o , o
o .0

O
I

 

O ' l
0' I,

0.0‘ 4'--—‘ '
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0

[Urease] (nM)

Figure 9. Quantitation of urease active sites.
Ureolytic activity at 37°C is shown as a function of enzyme
concentration following incubation at 0°C with PPD at 0.0 nM
(diamonds), 5.0 nM (circles), or 10.0 nM (squares), for 2.0
h (open symbols) or 96 h (solid symbols). Curves shown are
for data points fit to 2—h time points according to Williams

and Morrison (20).

87

urease studies, the pH dependence of boric acid inhibition
was consistent with trigonal B(OH)3 being the inhibitor, ra-
ther than tetrahedral B(OH);. Thiourea is not an inhibitor
of K. aerogenes urease despite its structural similarity to
urea. In addition, since no pH increase was observed when
urease was incubated in the presence of thiourea (100 mM),
it is probably not an alternative substrate either. Acryl-
amide, acetamide, thioacetamide and acetate exhibited no
significant inhibition at 500, 100, 20, and 20 mM respec-

tively.

DISCUSSION

As a framework for discussion of the K. aerogenes
urease inhibition results, we relate these studies to the
Zerner model (6) of the jack bean urease bi-nickel active
site, shown in Scheme 2. In the resting jack bean enzyme,
one nickel is suggested to coordinate a water molecule, and
a second nickel coordinates hydroxide. Urea displaces the
water molecule and the illustrated resonance structure is
thought to coordinate to nickel such that the positively
charged nitrogen is electrostatically stabilized by a nearby
carboxylate anion. A general base is proposed to activate
the nickel-coordinated hydroxyl group for nucleophilic at-
tack on the urea carbon to form a tetrahedral intermediate.

Decomposition of this intermediate and release of ammonia is

N 9:28

             

 

 

 

 

 

 

 

 

 

 

«12 f2/
:zC 6:2 :2 ouolokz
Ego I It V
I I I I
2.. B 68.6 a. I. a w
\ \oarm \
% 6.653.501
i /\
~12
n I . 1 la n I a ...6
12 as .m a 2 a... 12 a... .6 2
I Iz+ /\4 /
I l. / N N/ I I
.t . s .. . .4

 

 

 

 

89

thought to include general acid catalysis by a nearby thiol
group. Finally, carbamate is released with regeneration of
resting enzyme (6). Evidence for the active site carboxyl
and thiol groups in jack bean urease is primarily derived
from inactivation studies involving triethyloxonium ion and
alkylating agents (1).

If the above scheme is applicable to the bacterial
enzyme, ureolysis would require the participation of several
catalytic groups that may be sensitive to pH: a carboxylic
acid, a general base, a thiol, and the nickel-coordinated
hydroxide group. We have previously reported (7) that K.
aerogenes urease exhibited a.igax/Kh dependence on pH (over
the range pH 4.5-11) which was most easily interpreted by
assuming a catalytic requirement for a deprotonated active
‘site group with a pk; = 6.55 and a protonated group with a
pKh = 8.85. The Kb of urea is only moderately affected by
pH: i.e. a small decrease in K5 is observed upon protona-
tion of a group with pKh = 6.3 (Kh = 2.4 mM at pH 7-9 and K5,
= 0.72 at pH 5-6). The bacterial urease pH dependence of
catalysis is reasonably consistent with the Zerner model,

but specific group assignments for each pl; cannot be made.

Thiol Interactions with urease- Each of a series of
thiol compounds was shown to be a competitive inhibitor of
K. aerogenes urease. We have demonstrated that the thiolate
anions are the actual inhibitory species by examining the

effects of pH on cysteamine inhibition. In addition, we

90

used UV-visible spectroscopy to demonstrate direct inter—
action of the thiolate anions with the nickel center in bac-
terial urease, as evidenced by the development of charge
transfer bands. The demonstration of nickel-sulfur inter-
actions using a competitive inhibitor, where the spectrally
observed Kd is identical to the Ki: provides strong support
for a mechanism where urea also binds to the nickel metallo-
center, However, the data do not allow us to distinguish
which nickel (Nia or Nib) coordinates the thiolate. Support
for the presence of a negative charge at the urease active
site was obtained from the survey of different thiol inhi-
bitors (Table 1). Thiolates which possess a negatively
charged carboxylate group were significantly poorer inhibi-
tors than uncharged or positively charged thiolates. Thus,
the thiolate anion (R—S‘) inhibition is consistent with the
model shown in Scheme 3, where E-X‘ is an anionic active

site group.

Ni - -- ‘S-R Scheme 3

Scheme 3, derived from the bacterial urease thiolate
inhibition studies, has obvious similarities to portions of
the jack bean urease model (Scheme 2), where a carboxylate
is present at the active site. Using jack bean enzyme,
Zerner and co-workers (24, 28) also characterized the spec-
tral interactions associated with B-ME inhibition of jack

bean urease. Differences exist between the bacterial and

91

plant urease B-ME spectroscopic results, both in the absor—
ption peak positions (322, 374, and 432 nm versus 324, 380,
and 420 nm) and in the extinction coefficients at these
wavelengths (1115, 530, and 265 M'lcmrl vs 1550, 890, and
460 M’lcnn4), perhaps due to slight differences in the geo-
metry of the active site. (The extinction coefficients are
based on the concentration of active sites in order to allow

comparison.)

Phosphate Inhibition- Phosphate competitively inhibits
K. aerogenes urease in a pH dependent manner. The inhibi-
tion data are consistent with a requirement for protonation
of two ionizable groups with pk; values of <5 and 6.3.
These pk; values could be associated with the inhibitor,
with groups on the enzyme, or with both the enzyme and the
inhibitor; thus four simple mechanisms for phosphate inhi-
bition can be considered. In mechanism 1, both pK; values
are associated with phosphate, which is known to have pKfil
near 2 and pKa2 near 7 (these values can be affected by tem-
perature, ionic strength and buffer composition). In such a
mechanism the actual inhibitor is H3PO4 and the binding of
inhibitor is unaffected by the protonation state of the
enzyme. If H3PO4 were the true inhibitory species, a R; =
120 nM would be estimated. There is no obvious reason why
neutral H3PO4 should possess such high affinity for urease,
and it appears unlikely that an active site nucleophile

could attack H3PO4 to form a pentavalent intermediate at low

92

pH. Hence, we are disinclined to accept this mechanism. In
mechanism 2, both pk; values are associated with protonation
of enzyme groups, and the inhibitor is HZPog. K. aerogenes
urease is thought to possess an active site group with a pKh
= 6.55 (see above), which is in good agreement with the in-
hibitor pKh = 6.3. The protonation state of this active
site group changes the Kg for uncharged urea only 3-fold,
but may have a more significant effect on binding the neg-
atively charged inhibitor. Furthermore, the jack bean
urease active site model (Scheme 2) includes a carboxylate
which could correspond to the group with a pk; < 5. This
mechanism requires that the two phosphate pK; values be out-
side the pH range studied. Phosphate monoanion as the inhi-
bitor is consistent with the observed thiolate anion inhibi-
tion; however, it is unclear why two enzyme groups must be
protonated for highest affinity in the phosphate case. Per-
haps there is electrostatic repulsion of the delocalized
charge on H2P0; by two negatively charged groups at the ac-
tive site. In mechanism 3, the higher pKh is due to an
enzyme group and the lower pKh is due to phosphate. This
mechanism would require the pKhz of phosphate to be above 7
and the actual inhibitor to be H3PO4. As in mechanism.1
there is no compelling reason for H3PO4 to possess such high
affinity for the active site of protonated enzyme. Finally,
mechanism 4 assigns the lower pr to an enzyme group and the
higher to prz of phosphate. The actual inhibitor in such a

mechanism would be HZPOZ. As in mechanism 2 the phosphate

r L4

;

 

93

monoanion may bind in a manner similar to the thiolate
anions. The affinity is enhanced when a single enzyme group
with a pk; <5 (e.g. the carboxylate in the Zerner model is
protonated; electrostatic repulsion of the phosphate mono-
anion would then be decreased.

Since phosphate inhibition of urease is significant at
low pH values (e.gu at pH 6.5, K;=:6 mM for the total phos-
phate concentration), care must be taken when using phos-
phate buffer. Inhibition by buffer probably accounts for
previously reported irregularities in curves showing urease
activity as a function of pH (29, 30). Competitive inhibi—
tion of jack bean urease by phosphate was described as early
as 1949 (31); however, the pH dependence of inhibition has

never been examined for the plant enzyme.

Acetohydroxamic Acid Inhibition- Hydroxamic acids have
been reported to inhibit urease by competitive, non-competi-
tive, mixed, or irreversible mechanisms (Reviewed in Ref.
3). Most investigators, recognizing the time-dependent na-
ture of hydroxamate inhibition, have preincubated enzyme
with inhibitor for a fixed time and reported I50 values (in
the uM range), where 150 is the inhibitor concentration
which leads to 50% loss of activity. In many cases, it is
unclear whether equilibrium was reached during these pre-
incubation periods (e.g. to attain 95% equilibration of K.
aerogenes urease with 10 uM AHA requires approximately 3.5

h). In addition, I50 values do not provide any insight into

n?

5L

IQ.

94

the inhibitory mechanism. If equilibrium conditions were
reached and if the type of inhibition was known, then the
overall dissociation constant for inhibitor binding could be
related to the reported I50 values by taking into account
the substrate and enzyme concentration (32, 33). Our full
time course studies have resolved the mechanism of AHA inhi-
bition, concluding that AHA is a slow-binding competitive
inhibitor of K. aerogenes urease. Furthermore, we have de-
termined the rate constants for inhibitor binding and disso-
ciation, the ratio of which (hyis) is a useful measure of
efficacy when comparing inhibitors (16).

The initial.EJ complex may involve coordination of the
AHA anion (pk; = 8.7) to the metallocenter, analogous to
that proposed for other anions (e.g. thiols, phosphate). If
this mechanism is correct, one would expect K,- to be pH-de-
pendent, yet we observed no significant change in K; from pH
6.75 to 8.25. An alternative mode of AHA binding, similar
to the model proposed by Zerner and colleagues (6) for urea
binding (Scheme 2), would involve a resonance form of AHA
coordinating to the nickel and stabilized electrostatically
through interactions with a nearby active site anion (Scheme
4). The rate of 5.1 formation is probably not diffusion
limited; rapid E-I formation would be evident in progress
curve experiments (Figure 5) as changes in initial velocity
which depend on AHA concentration (16). Rather, we find
that at a particular urea concentration the initial veloci-

ties are the same (within 10%) for all AHA concentrations.

95

The slow transition which forms the E-I* complex may
arise from protein conformational changes or from inhibitor
coordinational changes at the active site. For example,
bidentate binding of AHA to the metallocenter may occur in
the enzyme as reported for hydroxamates with free metal ions
(36). An intriguing alternative is that AHA may bridge the
two nickel ions in the urease active site. For example,
nucleophilic attack by nickel-coordinated hydroxide may lead

to the following dead end complex.

EX‘ ax-
;H—OH NH—OH '
Ni..-0-é/ H—O‘---Ni = Ni--'O-C/—0' ,,,,, Ni Scheme 4
EH1 tn.
E1 El‘

Preliminary pH dependence studies of AHA inhibition are con-
sistent with such a model. AHA affinity decreases at lower
pH as would be expected if a catalytic nucleophile is re-
quired for complex formation. Also consistent with an event
involving the nickel center, the UV-visible spectrum of jack
bean urease is slowly perturbed upon AHA binding (34).
Zerner and colleagues have studied the kinetics of jack
bean urease inhibition by AHA (34, 35). Although they have
not demonstrated direct competition between AHA and urea,
the enzyme could not simultaneously bind B-ME and AHA (35).

Thus, they proposed the following mechanism:

S.,] 5.35.1“?

%

{Eigg Scheme 5
8 El

E

96

where (13 = 42 M’ls‘l, £4 = 76 x 10'55’1, and the equilibrium
dissociation constant K,- = 14/13 = 19.8 uM at pH 7.12 and
38°C (35). These kinetic constants for the plant enzyme
are similar to Its/K,- (35 M'ls‘l), its (8.7 x 1055’1), and K:
(2.6 uM) as determined for AHA inhibition of K. aerogenes
urease at pH 7.75 and 37° C, except that the value of I15 for
K. aerogenes urease is about 10-fold less than the corres-
ponding value of £4 for the jack bean enzyme. This in turn
decreases the K: for the bacterial enzyme about 10-fold com-
pared to the Kfcfifthe plant enzyme. The mechanisms of AHA
binding to bacterial and plant ureases appear to differ
(Scheme 1 versus Scheme 5). The distinction between the two
kinetic mechanisms arises from the ability to observe a lim—
iting apparent binding rate (ham) at high inhibitor concen-

trations using the bacterial enzyme. The i for AHA bind-

app
ing to jack bean urease was linear up to 8 mM AHA (35),
whereas we observed a hyperbolic response to increasing AHA
concentration with the K. aerogenes enzyme (Figure 3).
Zerner and colleagues (2) show a model in which AHA inhibits

the jack bean enzyme through bidentate coordination to one

nickel ion.

Phenylphosphorodiamidate Inhibition- PPD inhibition of
K. aerogenes urease also conformed to the slow-binding com-
*
petitive mechanism of Scheme 1; however, the K,- and Ki val-

ues were much smaller for PPD inhibition (95 nm and 94 pM)

97

than for AHA (1.5 mM and 2.6 uM). Previous studies of mi-
crobial urease inhibition by phosphoroamides were limited to
determination of 150 values, which were generally in the
nanomolar range (3).

In analogy to the postulated mechanisms for urea
(Scheme 2) and AHA (Scheme 4) binding, the E-I complex for

PPD may involve the structure shown in Scheme 6. The very

1' Y‘ E-X
+
NH HN NH,
1 \/
Ni--"O—P-NH_. H—O‘ ----- Ni :2 Ni.-‘O—P-O’ ----- Ni Scheme 6
i l
°\@ °@
E I E-I“

low K,- value for PPD (95 nM) may be a result of electro-
static stabilization and structural similarity to the tetra-
hedral intermediate postulated to occur during urea hydroly-
sis. The slow transformation which yields the very stable
E-I* complex may arise from nucleophilic attack by nickel-
coordinated hydroxide to form a pentavalent structure. As
in the case of slow-binding inhibition by AHA, the inhibitor
may bridge the two active site nickel ions. The parallel
nickel-bridging mechanisms forming the E-I' complex for AHA
and PPD are supported by the similarity of the £5 and £5 for
these inhibitors, although the PPD complex is more stable.
Two kinetic studies have been carried out analyzing the

effect of phosphoroamide compounds on jack bean urease

In

f‘

(I) 'T1

(T

98

according to Scheme 5 (28, 37). Phosphoroamidate (H3N-PO32‘)
was deduced to possess K;==1.9 mM and a dissociation rate
of 8.4 x 10'45‘1 at 25°C and pH 7.11 (28). These values do
not compare well with the bacterial urease PPD inhibition,
perhaps because of the highly anionic nature of phosphoro-
amidate. In contrast, N-acyl phosphoric triamides may in-
hibit jack bean urease in a manner similar to that observed
for PPD inhibition of K. aerogenes urease. A second order
rate constant for jack bean urease inhibition by methyl
butenyl phosphoric triamide was calculated to be 4 x 104
lfis’l (37). This rate compares reasonably well to the cor-
responding value for 115/K,- (7.8 x 105 M'ls’l) determined for
PPD inhibition of K. aerogenes urease. Furthermore, the
dissociation rates at 38°C, pH 7.01, of inhibitor-jack bean
urease complexes for PPD, methyl butenyl phosphoric tri-
amide, and phosphoric triamide were each 3.5 x.10‘5s‘1 (37).
This rate corresponds well to our value for'i5:= 4.7 x 10'5
5‘1. Because the three phosphoroamides yielded identical
dissociation rates in the jack bean study, the authors sug-
gested that each of these compounds is hydrolyzed by urease
to a common inhibitor, diamidophosphate (37). Although no
product analysis were presented in the jack bean urease
study, Zerner and colleagues propose a mechanism in which
the diamidophosphate product bridges the two nickel ions.
Since slowly processed substrates are known to behave kinet-

ically as slow-binding inhibitors for several enzymes (16).

VJ

99

we can not exclude PPD hydrolysis by K. aerogenes urease in

our experiments.

Boric and Boronic Acids--Boric and boronic acids were
earlier reported to be competitive inhibitors of P.
mirabilis urease (27) and now are shown to inhibit K.
aerogenes urease in a similar manner. The pH dependence of
inhibition was consistent with trigonal B(OH)3 being the in-
hibitor rather than tetrahedral B(OH);. As discussed else-
where (27), boronic acids are thought to inhibit several

proteases by reacting with an active site serine group.

 

Ser—OH + B(OH)3 > Ser—O—B(OH); Scheme 7

Thus, a reasonable scheme for urease inhibition by boric
acid is formation of the following complex, which is consis-
tent with the Zerner model involving nucleophilic attack by
hydroxide coordinated to nickel.

OH

HO - ‘B — O - Ni Scheme 8

OH
Such a complex may not be able to form a bridge between the
two nickel ions, thus borate does not exhibit slow-binding
kinetics. Analogous mechanisms involving nucleophilic at-
tack by nickel-coordinated hydroxide on an inhibitor could

also be envisioned for formation of the initial AHA and PPD

100

complexes. However, because AHA and PPD are analogs of
urea, we favor mechanisms shown in Schemes 4 and 6 for these
inhibitors.

In summary, the bacterial urease inhibition results are
consistent with most aspects of a model previously proposed
(6) for the active site of jack bean urease (Scheme 2). The
illustrated enzyme-inhibitor complexes are proposed as

working models to stimulate further studies.

kt)

'I.J

10.

11.

12.

13.

Andrews,

R. K.,

References

Blakeley,

R. L. & Zerner, B. (1984) in
Advances in Inorganic Biochemistry (Eichhorn, B. L.,

Marzilli, L. G., eds.) vol.

NY

Blakeley, R. & Zerner, B. (1984)
23. 263-292

Mobley, H. L. T. & Hausinger, R.
Microbial. Rev., (In Press)
Takashima, K., Suga, T. & Mamiya,

J. Biochem. 175, 151-165

Dixon, N.E., Gazzola, C.,
Zerner, B.

4133

Dixon, N. E., Riddles,
Blakeley, R. L., & Zerner,
Biochem. 58, 1335-1344
Todd, M. J., & Hausinger,
Chem., 262, 5963-5967
Mulvaney, R. L. & Bremner,

Biochemistry (Paul,

Vol. 5. pp.

Bremner,

Natl. Acad. Sci. USA,

P. W.,

6, pp. 245—283, Elsevier,

J. mol. Catal.

P. (1989)

G. (1988) Eur.

Blakeley, R. L., and
(1975) J. Am. Chem. Sac., 97 4131-

Gazzola, C.,

B. (1980) Can. J.

R. P.

J. M.

(1987) J. Biol.

(1981) in Soil

E. A. & Ladd, J. N., edS)

J. M. & Krogmeier, M. J.

153-196, Marcel Dekker Inc., New York

(1988) Proc.

85, 4601-4604

Mulrooney, S. B., Pankratz, H. S.
(1989) J. Gen..Micrabial. 135,
Weatherburn, M. W. (1967) Anal.
971-974

Lowry, O. H., Rosebrough, N. J.,
Randall, R. J. (1951) J. Biol.
275

Lineweaver, H. & Burk, D. (1934)
Sac., 56, 658-666

101

& Hausinger, R. P.
1769-1776 -

Chem., 39,

Farr, A. L. &

Chem., 193, 265-

J. Am. Chem.

&

WC

N;-

44

«Ifu

Ale

’\

«1C

3 i

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

102

Wilkinson, G. N. (1961) Biochem. J., 80, 324-
332

Dixon, M., and Webb E. C. (1979) in Enzymes,
Academic Press Inc., New York

Morrison, J. F. & Walsh, C. T. (1988) Adv.
Enzymal., 61, 201-301

Schreiner, W., Kramer, M., Krischer, S. &
Langsam, Y. (1985) PC TECH J., 3, vol. 5, 170-
190

Marquardt, D. W. (1963) J. Soc. Indust. App.
Math., 11, 431-441

Williams, J. W. & Morrison, J. F. (1979) Math.
Ehz., 63, 437-467

Danehy, J. P. & Noel, C. J. (1960) J. Amer. Chem.
Sac., 82, 2511-2515

DeDekon, R. H., Broekhuysen, J., Bé Chet, J. & Mortier,
A. (1956) Bichiama Biophysica Acta 19, 45-52

Sjoberg, B. (1942) Berichte der Deutschen Chemischen
Gesellschaft, 75, 13-29

Friedman, M., Cavins, J. F. & Wall, J. S. (1965)
J. Amer. Chem. Sac., 87, 3672-3682

Blakeley, R. L., Dixon, N. E. & Zerner B. (1983)
Biochim. Biophys. Acta, 744, 219-229

Hase, J. & Kobashi, K. (1967) J. Biochem.
(Tbkya) 62, 293-299

Dixon, N. E., Gazzola, C., Watters, J. J.,
Blakeley, R. L. & Zerner, B. (1975) J. Am.
Chem. Sac., 97, 4130-4131

Breitenbach, J. M. & Hausinger, R. P. (1988)
Biochem J., 250, 917-920

Dixon, N. E., Blakeley, R. L. & Zerner, B.
(1980) Can. J. Biochem., 58, 481-488

Kamel, M. Y. & Hamed, R. R. (1975) Acta Biol.
.Med. Germ., 34, 971-979

Nakano, H., Takenishi, S. & Watanabe, T. (1984)
Agric. Biol. Chem., 48, 1495-1502

 

 

31.

32.

33.

35.

36.

37.

38.

103
Harmon, K. M. & Niemann, C. (1949) J. Biol.
Chem., 177, 601-605
Cha, S. (1975) Biochem. Pharm., 24, 2177-2185

Cheng, Y. C. & Prusoff, W. H. (1973), Biochem.
34Pharm. 22, 3099-3108

Blakeley, R. L., Hinds, J. A., Kunze, H. B., Webb, E.
C. & Zerner, B. (1969) Biochemistry, 8, 1991-2000

Dixon, N. E., Hinds, J. A., Fihelly, A. K.,
Gazzola, C., Winzor, D. J., Blakeley, R. L. &
Zerner, B. (1980) Can. J. Biochem., 58, 1323-
1334

Neilands, J. B. (1967) Science, 156, 1443-1447

Andrews, R. K., Dexter, A., Blakeley, R. L. & Zerner,
B. (1986) J. Am. Chem. Sac., 108, 7124-7125

REACTIVITY OF THE ESSENTIAL THIOL
OF KLEBSIELLA AEROGENES UREASE:

EFFECT OF DH AND LIGANDS ON THIOL MODIFICATION

ABSTRACT

The kinetics of Klebsiella aerogenes urease inactiva-
tion by disulfide and alkylating agents was examined and
found to follow pseudo-first-order kinetics. Reactivity
of the essential thiol is affected by the presence of sub-
strate and competitive inhibitors, Consistent with a cys-
teine located proximal to the active site. In contrast to
the results observed with other reagents, the rate of ac-
tivity loss in the presence of 5,5'-dithiobis(2-nitroben-
zoic acid) (DTNB) saturated at high reagent concentra-
tions, indicating that DTNB must first bind to urease be-
fore inactivation can occur. The pH dependence for the
rate of urease inactivation by both disulfide and alkyla-
ting agents was consistent with an interaction between the
thiol and a second ionizing group. The resulting macro-
scopic pig'values for the two residues are <5 and 12.
Spectrophotometric studies at pH 7.75 demonstrated that
2,2'-dithiodipyridine (DTDP) modified 8.5 i 0.2 mol
thiol/mol enzyme, or 4.2 mol thiol/mol catalytic unit.

With the slow-tight binding competitive inhibitor

104

105

phenylphosphorodiamidate (PPD) bound to urease, 1.1 i 0.1
mol thiol/mol catalytic unit was protected from
modification. PPD-bound, DTDP—modified urease could be
reactivated by dialysis, consistent with the presence of
one thiol per active site. Analogous studies at pH 6.1,
using the competitive inhibitor phosphate, confirmed the
presence of one protected thiol per catalytic unit. Under
denaturing conditions, 25.5 i 0.3 mol thiol/mol enzyme

(Afi.=:211,800) were modified by DTDP.

I“ 7“

r‘4

INTRODUCTION

Urease [urea amidohydrolase (EC 3.5.1.5)] is a
nickel-containing enzyme that catalyzes the hydrolysis of
urea to ammonia and carbamate; the carbamate spontaneously
hydrolyzes to form a second molecule of ammonia and car-
bonic acid (2 15). The enzyme from the plant Canavalia
ensifarmis (jack bean) is the best characterized urease;
however, the role of the plant protein is unclear (29).
Dixon et al. (6) proposed a model for the hydrolysis of
urea by jack bean urease in which one nickel coordinates
the oxygen atom of urea, polarizing the carbonyl group,
and a second nickel coordinates hydroxide ion, the cata-
lytic nucleophile. Three amino acid residues were sug-
gested to participate in catalysis: a carboxyl group, a
sulfhydryl group, and an unidentified base. Consistent
with the presence of an essential thiol group, both alk-
ylating and disulfide reagents were shown to inhibit the
enzyme (1, 7, 19). The essential thiol exhibited a pk; of
9.15, and at pH 7.7 reacted more slowly than several other
thiols which were not required for activity (19).

In contrast to the jack bean enzyme, very little is
known about the thiol reactivity of microbial ureases.
Characterization of the bacterial enzyme is essential be-

cause of its important role in cellular nitrogen

106

107

metabolism and certain pathogenic states (reviewed by
Mobley and Hausinger, 15). Ureases from Brevibacterium
ammoniagenes (18), Ureaplasma urealyticum (22), and a bo-
vine ruminal population (13) have been shown to be inhi—
bited by alkylating reagents and p-chloromercuribenzoate;
however, the kinetics of activity loss and other proper-
ties of the reacting group were not analyzed.

We demonstrate that Klebsiella aerogenes urease has
an essential thiol and examine the pH dependence of its
reactivity with alkylating and disulfide reagents. We
also examine the effect of substrate and several competi-
tive inhibitors on the rate of inactivation to assess
whether this essential thiol is at the active site.
Finally, we characterize spectroscopically the reactivity
of both essential and non-essential thiol groups in

urease .

EXPERIMENTAL PROCEDURES

Enzyme purificatian- K. aerogenes urease was purified
to a specific activity >2500 umol-mindomgd-from strain CG
253 carrying the recombinant plasmid pKAU19 (17) by using
procedures previously described (26). For spectroscopic
analysis of urease thiols, the final enzyme purification
step was carried out with buffers containing DTT rather

than B-ME. One unit of urease activity is defined as 1

.1”,

v...»

o..‘
9""

at

e..-

U’
(I)

‘ve

Lee

‘1.

.M

(I)

V~

‘v

.‘

 

 

:-
L

108

umole urea degraded per minute in the standard assay mix-
ture, which contained 50 mM urea, 5 mM EDTA, 50 mM HEPES,
pH 7.75 at 37°C. Linear regression analysis of the re-
leased ammonia, determined by conversion to indophenol
(28), versus time yielded the initial rates. Protein was

assayed by the method of Lowry (12).

Kinetics of urease inactivation- Inactivation re-
actions contained 80-100 mM buffer, 8 mM EDTA, plus the
indicated reagent concentrations. All thiol modification
reagents were dissolved in H53, except DTDP and DTNB,
which were dissolved in 100% HPLC grade methanol. Al-
though low concentrations of methanol were found to have
no apparent effect on urease activity, reaction series
using these reagents were kept at a constant concentration
of methanol, usually <2%. Inactivation reactions were
initiated by addition of a small amount of enzyme (<1% by
volume) and aliquots were periodically diluted 50-fold
into the standard assay for activity determination. This
dilution effectively halted any further inactivation
during the assay as shown by the linearity of the initial
rate determination. Pseudo-first-order rate constants
were obtained from plots of ln(percent activity remaining)
versus time. The pH of the inactivation reactions was
taken immediately following completion of the assays.

Progress curves in the presence of urease inactiva-

tors were typically carried out using 5 ml reactions

 

109

containing 8 mM EDTA, 80 mM HEPES, pH 7.75 at 37°C, and
0.10 ml aliquots were analyzed periodically for ammonia.
Experiments in the absence of urea demonstrated a pseudo-
first-order loss of initial velocity consistent with

Vt : V0 641,th

where I; = initial velocity of urease in the absence of
modifying reagents, R = concentration of modifying re-
agent, and hum = pseudo-first-order rate constant for loss
of urease activity. Pseudo-first-order rate constants in
the presence of urea were obtained by fitting the product,

ammonia, produced as a function of time (Pfi to:

vb

app

(1 .. e—kappt ) (2)

 

PI=P0+k

where P0 = product at time t:= 0. Except for DTNB, these
rate constants were proportional to the concentration of
inactivating reagent at all urea concentrations tested,
showing that loss of activity is a second-order process,
even in the presence of saturating concentrations of sub-

strate (50 mM urea).

Kitz and Wilson (9) have shown that the rate of amino
acid modification in the presence of ligands can be repre-

sented by:

(I)

(#7

'1

('3

rr1

Ill

(7

'TI

rr

110

l

'< I
E +F? -————€>Inoc1we>enzygm9

2
E L + R ——> Inactive enzyme

(Scheme 1)

where E = enzyme, R = modifying reagent, L = ligand, (11 =
second-order rate constant for the inactivation of free
enzyme, ha: second-order rate constant for inactivation
of ligand—bound enzyme, and Ki = equilibrium dissociation
constant for ligand binding. If the concentration of
modifying reagent is much greater than the concentration
of enzyme, the effect of the ligand on the apparent

pseudo-first-order rate constant, hum, will be given by:

it2
(KL+ ,1 [L])

 

 

[R] I, <3)

app = KL + [L]

such that iapp will be dependent on the ratio of £2 to £1:
if £2 > (11, the ligand increases kapp; and if (12 < 111, then
the ligand decreases hxm' providing complete protection
of the susceptible amino acid as (12 approaches zero.
Pseudo-first-order rate constants for urease inactivation
in the presence of ligands were fit to equation 3 using

the Markfit program (see below).

111

Reactivation of DTNB— or DTDP—inactivated urease-
Urease was incubated with 2 mM DTNB or with 0.5 mM DTDP
for 15 min, then diluted ZOO-fold into 5 mM EDTA, 50 mM
HEPES, pH 7.75 at 37°C containing various concentrations
of DTT. Aliquots were periodically diluted into the
standard assay mixture to measure the activity. Initial
velocity data.(l§) were fit to the following expression

for the pseudo-first-order recovery of activity:

VI: V0 + Vmax (l - e-kappt) (4)

where V; = initial velocity of DTDP- or DTNB—modified
urease, Vmam = maximum velocity recovered, and kapp =

apparent pseudo-first-order rate constant for the

reactivation of disulfide-inactivated urease by DTT.

Spectroscopic analysis- Phosphate and dithio-
threitol, used to stabilize urease during storage, were
removed by ultrafiltration using an Amicon Centricon-3O
microconcentrator (5 X 1:10 dilution). Spectroscopic data
were obtained by using a Gilford Response spectrophoto-
meter with l.O-cm.microcuvettes, at 37'C, assuming that 2-
thiopyridine had 6343 mm = 7030 M'lcm‘1 (8) . Independent de-
terminations of 5u3rm were done using each of the buffer
solutions by two methods: aliquots of DTDP were added to
cuvettes containing 1 mM B-ME or 1 mM cysteine, or ali-

quots of B-ME were added to cuvettes containing 0.5 mM

112

DTDP. Identical results were obtained by both methods,
confirming the value of 7030 for 6M3rm of 2-thiopyridine.
The total number of urease thiols was determined by re-
acting urease with DTDP (0.5 mM) in the presence of 5.5 M
guanidine-HCl, 10 mM EDTA, 0.1 M Tris-HCl, pH 7.9. The
maximum absorbance at 343 nm was attained within 30
seconds and did not change after an additional 30 minutes.
The concentration of protein was determined by the method
of Lowry (12), and the molar concentration of urease was
calculated based on.Afl.= 211,800 (16).

The rates of reaction for the accessible thiols in
native enzyme were assessed by scanning three cuvettes for
each reaction: a buffer blank, DTDP control, and DTDP +
6—8 uM urease. Spectral data were transferred to an IBM
PC to correct for the absorbance increase due to spontan—
eous degradation of the thiol reagent (<O.1% spontaneously
hydrolyzed in 30 minutes at pH 9.0). During the time
scan, aliquots were removed for assay using the standard
assay mixture. For all reactions involving native enzyme,
DTDP was added last, allowing ample time for enzyme and
competitive inhibitor (when present) to equilibrate.

Prior to DTDP addition, activity measurements of urease

and AHA or urease and PPD showed >99.9% inhibition.

Curve fitting- Curve fitting made use of a BASIC

program termed Markfit (23) which employs the Marquardt

113

algorithm (14) to fit by non—linear least squares a set of

data points to a rate equation.

RESULTS

Inactivation of urease by thiol-specific reagents-
Homogeneous K. aerogenes urease exhibited a pseudo—first-
order loss of activity in the presence of disulfide and
alkylating reagents. As illustrated for DTDP, pseudo-
first-order rate constants were directly proportional to
the concentration of inactivating reagent (Figure 1)
allowing calculation of the apparent second—order rate
constant. Similar results were obtained for several other
reagents, and their second—order rate constants for the
loss of ureolytic activity are provided in Table 1.

Inactivation of urease by DTNB is also a pseudo—
first-order process; however, the apparent rate constant
reaches a limiting value at high concentrations of DTNB.

Preliminarily, the data were fit to the following model.

k k
E + R TIE—i E R 4% inocTive enzyme
2 (Scheme 2)

The apparent pseudo-first-order rate constant for an

inactivator that exhibits a binding step follows

114

 

c1
-(
d
-:
fl
-1
q
.1
4
.—
.1
.4
.1
.,
_
.4
q
q
q

/
J.

0.05«

Ln <

003-4

Low (9")

0.01-
-2.0— , r
" O 1 2
‘ DTDP (mM)

 

 

01-1
b

I
_A
O
l
.
111111111i1111141111111

 

 

 

T T r T r Y T —l T Y I I T f T I Y I

0.0 0.5 1.0 1.5 2.0

Tkne (nfinutes)

FIGURE 1. Pseudo-first-order urease activity lose
in the presence of DTDP. The fraction of activity remain-
ing is shown as a function of time for urease [4.4 nM],
reacting with DTDP at 0 (V), 0.25 (O), 0.5 (I), or 1.0
mM (A). All reactions were in 8 mM EDTA, 80 mM HEPES at
pH 7.75, and 37'C. INSET: Determination of the second-

order rate constant for urease inactivation by DTDP.

115

Table 1

Second—order rate constants for loss of urease activity

by thiol-specific reagentsa

 

 

Inactivator Concentration Rate constant
examined 1 Standard error
awn) (Mfls-1>
Alkylating
reagent
IAA 10-25 0.002
IAM 1-200 0.026 1 0.001
NEM 0.050—0.200 4.76 i 0.19
Disulfide
reagent:
DTNBb 0.1-5.0 7.2
DTDP 0.01—4.5 22 i 5
Cystine 0.5-1.0 0.077 t 0.004
Cystamine 0.25-1.0 27 t 1
MMTS 0.025-0.100 710 t 50

 

a

Values determined in 8 mM EDTA, 80 mM HEPES, pH 7.75
buffer at 37'C.

Inactivation of urease by DTNB is not a simple second
order reaction. The pseudo-first-order rate can be sat-
urated at high concentrations of DTNB (Scheme 2);
therefore, the apparent second-order rate constant shown
15

M3

k = k2

app

116

 

i. (3&1 R
app _ £1 R + £2 + £3 (5)
under steady state conditions (9). Since the inactivation

step occurs much slower than the rapid equilibrium between

enzyme and modifying reagent, L3<a:&2, and

1 1 (2
I ' £3 + 9.17.312

 

(6)

Therefore under steady state conditions, double reciprocal

plots of 1/£ versus 1/[R] yield a straight line. The

app
results for DTNB inactivation of urease fit this pattern
as shown in Figure 2. Values for £3 and liz/Iil = K, were
calculated to be 0.0065 s—-1 and 0.9 mM, respectively.

To test whether thiol modification released the
tightly bound nickel ions of urease, enzyme modified by
IAM, DTDP, or MMTS was dialyzed to remove any low molecu-
lar weight compounds, then subjected to atomic absorption
analysis as previously described (25). Enzyme samples

which had lost >90% activity retained essentially 100% of

the original amount of nickel.

Reactivation of urease inactivated by disulfide re-
agents- Urease which was inactivated by DTNB slowly
recovered activity after dilution into buffer containing
DTT, as illustrated in Figure 3. Data for reactivation at

three DTT concentrations were fit as described in

 

117

 

1300 d I 1 I r I V T I F I I Y T V I I I' I V l 1

be

99

 

 

 

 

FIGURE 2. Saturation of the apparent rate of urease
modification by DTNB: the effects of urea and borate.
Pseudo-first—order rate constants for the modification of
urease by DTNB were obtained from plots similar to Figure
1. (app-1 (54), obtained at pH 7.75 in 8 mM EDTA, 80 mM
HEPES, 37°C, is shown as a function of [DTNB]4-(HWF1) for
urease (I), urease in the presence of 2 mM (A) or 5 mM

(I) urea.

 

 

118

 

>‘
;:
o
52 T
o ai—
> 67E
“E’EE
>. E
N
c
u)

 

 

050
[DTT] (mM)

 

O 25 50 75 100 125

Thne (nfinutes)

FIGURE 3. Reactivation of DTNB-inactivated urease.
Urease (17 nM), previously inactivated by DTNB, was in-
cubated in the presence of 8 mM EDTA, 80 mM HEPES pH 7.75
with o (O), 0.1(0), 0.15 (I), or 0.2 (A) mMDT‘I‘ at 37'C.
The initial velocity was determined at the indicated times
by using the standard assay. INSET: Determination of the
second-order rate constant for the reactivation of DTNB-

modified urease by DTT.

 

ll9

Experimental Procedures, and the second-order rate
constant for urease reactivation (3.1 i 0.2 Ethyl) was
calculated from regression of the slope of kum versus DTT
concentration. Similar DTT reactivation results were
obtained for enzyme inactivated with DTDP (second—order
reactivation rate constant 2 2.7 i 0.3 M4snfl.

In contrast to the recovery of activity in the
presence of DTT, no activity was recovered when DTDP—
modified enzyme was treated with excess KCN (500 mM) for
120_minutes at 37'C. Cyanolysis did occur as shown spec—
trophotometrically by the release of 4 mole 2-thiopyri-
dine/mole enzyme. Addition of DTT (1 mM) to cyanide-
treated DTDP—inactivated urease did not release additional

2—thiopyridine.

Effect of ligands on the rate of thiol modification-
Having shown that thiol-reactive reagents inactivate K.
aerogenes urease, experiments were designed to test
whether ligands which bind to the active site of urease
alter the rate of inactivation.

The effect of two competitive inhibitors of urease on
the rate of inactivation by IAM is shown in Figure 4.
Phosphate was previously shown to be a competitiVe inhibi-
tor of urea hydrolysis, with a K}that depended on pH in a
complex manner (26). Since phosphate binds with higher
affinity to urease at lower pH, the effect of phosphate on

urease inactivation was examined at pH 6.1 in MES buffer

 

120

100

 

 

 

kopp (2'. control)

20a

 

 

 

 

[Competitive inhibitor] (mM)

FIGURE 4. Effect of the competitive inhibitors boric
acid and phosphate on the rate of urease inactivation by
IAN. The second-order rate constants for urease inactiva-
tion by IAM were determined in the indicated concentra—
tions of phosphate (I) and boric acid (0), and are ex—
pressed as percent of the inactivation rate in the absence
of inhibitors. Reaction conditions were 8 mM EDTA, 80 mM
buffer (MES, pH 6.1 for phosphate inhibition, HEPES pH 7.7
for borate inhibition) at 37'C.

121

where the Kfi,mmqmm£) = 1 mM. Increasing concentrations of
phosphate decreased the apparent urease inactivation rate;
at saturating phosphate concentrations hum was less than
2% that of the control. Increasing concentrations of
boric acid also reduced the apparent rate of inactivation;
however, saturating levels of borate decreasedfiéapp by
only two-thirds. This difference in behavior between
phosphate and borate is not due simply to a pH effect,
because borate slowed the rate of urease inactivation by
IAM to the same extent at pH 6.0, 7.75, and 9.5 (data not
shown).

Values for &y%1 (Scheme 1), the ratio of inactivation
rates for ligand-bound and free enzyme, and Ki, the ligand
dissociation constant, were calculated from the above
studies. These values are provided in Table 2, which also
lists values obtained in analogous IAM inactivation exper-
iments using the inhibitors BBA and 4-Br-BBA and values
for the inhibition of urease inactivation using the di-
sulfide reagent DTDP. While 4—Br-BBA parallels borate in
providing partial protection of the essential thiol, BBA
has little effect on urease inactivation by IAM, and ac-
tually increased the rate of urease inactivation by DTDP
(Table 2). The effect of the ligand urea on the rate of
urease inactivation by DTDP was assessed by analysis of
progress curves. Urea also provided partial protection of
the essential thiol (Table 2), with the apparent second-

order rate constant decreasing about 2.5-fold.

122

Table 2

Effect of Urease Ligands on the Rate of Inactivation by

Thiol-Specific Reagentsa

 

 

 

 

Inactivator
IAM DTDP
RQioriKm
for urea
Ligand hydrolysis flab Ki E2 Ki
(mM) 7E1 (mM) "1 (mM)
Borate 0.34 0.32 0.30 0.49 0.45
BBA 10 0.8 11 2.6 13
4-Br-BBA 0.37 0.24 0.34 n.d.C ndd.
Phosphated 0.8 <0.02 0.50 <0.02 0.43
(pH 6.1)
ureae 2.5 n.d. n.d. 0.4 3

 

a All reactions were done in 8 mM EDTA, 80 mM buffer (pH
6.1, MES; pH 7.75, HEPES) at 37°C. The pseudo-first-
order rate of activity loss was measured as a function
of inhibitor concentration, then rate constants were
fit to equation 3.

b £2 and 51 are defined in Scheme 1.

n.d. not determined.

6 Initial attempts to fit phosphate data to equation 3
showed (a2 less than 2% (c1, so KL was obtained by fixing
‘12 = 0.

e The pseudo-first-order rate for activity loss in the
presence of urea was obtained from progress curves in 8
mM EDTA, 80 mM HEPES, pH 7.75, at 37°C and various con-
centrations of DTDP. Pseudo-first-order rate constants
were obtained by fitting data to equation 2, then these

pseudo-first-order rate constants were fit to equation
3.

123

Ligands were also examined for their effects on the
rate of activity loss in the presence of DTNB, which was
shown above to exhibit saturation of the apparent rate at
high reagent concentrations, consistent with Scheme 2.
Progress curves for ureolysis in the presence of DTNB
yielded rates for activity loss which when plotted as
(app4 vs [DTNB]4, were parallel. Saturating concen—
trations of substrate decreased the maximum rate, £3, 10—
fold, and also decreased the concentration of DTNB re—
quired to reach half the maximum rate, kydl, by 10-fold.

The effect of pH on t Figure 5 illustrates the

app”
effect of pH on the rate of K. aerogenes urease inactiva—
tion by IAM and DTDP. Above pH 12, the rate does not ap-
pear to be affected by changes in pH. Between pH 8.5 and
12, plots of log ham vs pH have a slope of +1. Between

pH 8 and 5, the rate of modification is not affected by pH
changes; lower values of pH led to rapid denaturation of
urease. The pH dependence for DTNB inactivation is comp-
licated because of the saturation observed at high DTNB
concentrations (as described above). In terms of the re-
action illustrated in Scheme 2, the fig/£1 (or K.) of DTNB
appeared to be nearly independent of pH, whereas £3 for
DTNB modification of urease behaved similarly to the other

inactivators (Figure 6). Inactivation by DTNB was exam-

ined only below pH 9.5 due to reagent instability.

124

 

 

5:
Z 0:3
5': 8
I v
4-3 s
”a 32 e
8- : .
Q 23 o ..
V : DTDP ° '
g 1€OegfleuA :V
—l : :1
O—: O,
12 .1
:. IAM I ‘3
:Q..I-AA
_2 IIIIIIIIIIIII'IFI]
5.0 7.0 9.0 11.0 13.0

pH

FIGURE 5. The pH dependence of IAM and DTDP inac-
tivation of urease. The log of the apparent second-order
rate constants in (Mfihrl) are shown as a function of pH
for DTDP (open symbols) or IAM (closed symbols) for reac-
tions carried out in 8 mM EDTA, 100 mM buffer: acetate
(circles), MES (squares), HEPES (triangles), CHES
(diamonds), CAPS (inverted triangles), or phosphate

(circles).

125

 

I I T I I I T I I I I I I T I I I I I I I I I I I I I T

I
N
O

1

 

|
P’
01
[L441
:1
I
I

l

1

Log(Ki or ‘4)

L l

(L.
C)
.1
h
I

T

 

11
I
I

l

I
9
U1

1

1
II
lllllLllllllllllliJllllllJiiJllll

111

 

 

 

I
P
C)

T I I I I I I I T I I T I I l T I I T I 1T7 T I I T T I I

‘10 ‘15 8f) 85 90 95
pH

9’
(n

FIGURE 6. Effect of pH and DTNB concentration on the
pseudo-first-order rate of urease activity loss. The rate
of urease inactivation by DTNB was determined in 8 mM
EDTA, 80 mM buffer, at 5 or more concentrations of DTNB at
37’C using MOPS, HEPES, TAPS, and CHES. K,- and k3 (Scheme
2) were determined by non-linear least squares analysis of
the pseudo-first-order rate constants versus DTNB concen-

tration; log K,- (M) ( O ) and log (c3 (5'1) ( I ) are shown

as a function of pH.

126

Spectrophotometric quantitation of urease thiols- In
addition to the essential thiol group, other cysteine res-
idues may react with disulfides or alkylating agents without
loss of urease activity. The total number of urease thiols
and the number and reactivity of accessible thiols in native
enzyme were probed by Spectrophotometric methods.

DTT treated urease (free from low molecular weight
thiols) was incubated with DTDP in the presence of 5.5 M
guanidine. The number of thiol equivalents was divided by
the molar protein concentration to demonstrate that 23.2 i
0.3 mol thiol/mol denatured enzyme could react (assuming the
native molecule possesses an 0125474 stoichiometry with Mr =
211,800).

Figure 7a shows progress curves for the reaction of
DTDP with urease at pH 7.75. The release of 2-thiopyridine
corresponded to the reaction of 8.5 i 0.2 mol thiol/mol
enzyme. The DTDP—reactive cysteines were converted to mixed
disulfides with 2-thiopyridine, rather than to cystine res-
idues, as shown by DTT—dependent release of > 8.3 i 0.3 mol
2-thiopyridine/mol of the isolated, inactivated enzyme with
concomitant recovery of activity. When urease was equili-
brated with the slow, tight-binding competitive inhibitor
PPD prior to reaction with disulfide reagents, only 6.3 i
0.4 mol thiol/mol enzyme reacted. Urease modified in the
presence of PPD recovered over 70% of its original activity
following prolonged dialysis versus 1 mM EDTA, 50 mM phos-

phate, pH 6.5, at 4°C, (at 1:20,000 volumes, with several

127

 

.e
O

W
I!" +5?“MAMA

/

lllLlllllAll

\

/ +5opMPPD

lllllll I

ll]

Thiols modified at pH 7.75
N U ¥ 0'! O \J D D

an.

 

IIIJJAILLAL AllllllllLJAllllll

 

 

 

._
(D
I
Q' :
"" .2
o d
u :
.9 r
3‘: '1
U 1
° 1
E :
.2! difference _
.9 1
.c .
F‘ :
o I I V V I r T U T V r Y T T T I T T V U r T I T r r T Y—r T
O 5 1 O 1 5 20 25 30

Time (minutes)

FIGURE 7. Spectroscopic progress curves for the re-
action of urease with DTDP. Absorbance at 343 nm was con-
verted into the number of urease thiols reacting and is
shown as a function of time. DTT-treated urease (6-8 HM)
was reacted with 0.5 mM DTDP in 8 mM EDTA, 80 mM buffer
(HEPES, pH 7.75 or MES pH 6.1) in a 1.0 cm micro-cuvette.
Samples include: A) urease at pH 7.75, urease equili-
brated with 10 mM AHA or 50 uM PPD, and the absorbance
difference between control and PPD inhibited urease, and
B) urease at pH 6.1, urease equilibrated with 10 mM
phosphate, and the difference scan.

128

buffer changes). In contrast to PPD, in the presence of the
slow binding inhibitor AHA, 8.1 i 0.1 mol thiol/mol enzyme
were accessible to modification by DTDP, and prolonged di-
alysis resulted in recovery of <2% the original activity.
Figure 7b illustrates the DTDP reactivity of urease at pH
6.1 in the presence and absence of phosphate. Phosphate
protects 2.0 i 0.2 mole thiol/mole enzyme from modification
by DTDP. An initial rapid phase was followed by a slow,
steady increase (perhaps due to low pH—induced
denaturation), so that after 30 min, 8.5 i 0.5 urease thiols

had reacted.

Figure 8 compares the percent activity of urease as a
function of the number of urease thiols modified at pH 6.1
and 7.75 according to the method of Tsou with 1: 1 (27).
Tsou plots with c> 1 did not fit the data and are not
shown for clarity. The essential thiols reacted more
rapidly than non-essential thiols at pH 6.1, possibly as a
result of the plateau observed in the pH dependence
studies. The slowest phase of urease modification by DTDP
at pH 6.1 is not correlated to modification of the essen-
tial thiol, as > 95% of the activity is lost by the time 5

thiols have reacted.

129

 

 

Fraction Activity Remaining
O
03
l

 

 

 

 

# Thiols modified

FIGURE 8. Activity as a function of urease modifica-
tion. At periodic intervals during the scans in Figure 7,
8.5 pmoles of urease was withdrawn to assay under standard
conditions. Fraction activity remaining (aU‘ where 1: 1)
is shown as a function of total number of thiols modified
at pH 6.1 (I ), and pH 7.75 ( I ), according to the
method of Tsou (27).

DISCUSSION

Demonstration of an essential thiol in K. aerogenes
urease- Reaction of urease with either disulfide reagents
or alkylating reagents led to at least a 10,000-fold de-
crease in activity, consistent with the presence of at
least one essential enzyme thiol. The decrease in activ~
ity is probably not a steric effect, since cyanolysis of
the bulky 2-thiopyridine group to yield the small un-
charged cyanide derivative failed to increase ureolytic
activity. Similarly, modification of urease thiols with
MMTS, resulting in the small, uncharged methylthio- deriv—
ative, led to the loss of >99.9% of activity. Inactiva-
tion was not accompanied by a loss of nickel ion, as shown
by atomic absorption analysis. Furthermore, DTT-dependent
release of either 5-thio-2-nitrobenzoate or 2—thiopyridine
led to 100% recovery of the initial activity.

Comparison of the second-order rate constants for
urease inactivation by the eight thiol-modifying reagents
in Table 1 indicates that reagents containing anionic
groups react with urease much more slowly than correspond-
ing neutral reagents, and the positively charged cystamine
reacts over 300-fold faster than the zwitterionic cystine.
These observations are consistent with electrostatic ef-

fects caused by an anionic group proximal to the essential

130

131

thiol. We previously suggested the presence of an anionic
group at the active site of K. aerogenes urease based on
comparison of the ngalues for a series of competitive
inhibitors (26). These two observations may be related if

the essential thiol is at the urease active site.

Is the essential thiol at the active site? Loss of
urease activity by thiol-specific reagents may be due to
modification of an active site thiol, or to modification
of a non-active site residue, resulting in a conforma-
tional perturbation. In an attempt to discriminate be—
tween these two possibilities, the effects of urease ac-
tive site ligands on the rate of inactivation was ex-
plored.

Four competitive inhibitors of urease and the enzyme
substrate, urea, were shown to affect the rate of enzyme
inactivation, as summarized in Table 2. Whereas urea,
boric acid, and 4-Br—BBA afforded only partial protection
of the essential thiol, phosphate completely protected the
enzyme against inactivation by either IAM or DTDP. In all
cases, the effects were achieved at concentrations cor-
responding to K}cm'K; of the ligand. These results are
consistent with the essential thiol being at the urease
active site. Since urea provides only partial protection
from inactivation, the essential thiol may not be directly
involved in urea binding, but could participate in catal-

ysis. These results are consistent with a model for the

132

jack bean urease active site in which a thiol is proposed

to serve as a proton donor (6).

.pH dependence of urease inactivation by IAM'and DTDP-
Roberts et a1. (21) have shown that the intrinsic react-
ivity of a thiolate anion with MMTS is >5 X 109 times that
of the protonated thiol. Similar results were obtained by
Bednar (3) for the reaction of thiols with NEM. Thus, the
apparent pseudo—first-order rate constant for either di-
sulfide exchange or alkylation can be calculated by

Kafi’l
"app z [3+] + K“

 

(7)

where %.15 the second-order rate constant for reaction
with the thiolate anion, and K; is the acid dissociation
constant of the amino acid which is modified. A plot of
log(h¥m) versus pH for thiols exhibiting this behavior
would possess a slope of 1 below the pK and a slope of 0
above the thiol pKL The pH dependence of urease inact—
ivation by IAM and DTDP is inconsistent with this simple
mechanism; La the reactivity of the essential thiol is pH
independent below pH = 8 for both reagents in Figure 5.
One mechanism to explain the data in Figure 5 invokes
a greatly enhanced reactivity of protonated reagent to-
wards the enzyme thiolate anion. Brocklehurst et al. (4)
demonstrated that protonated DTDP (pK = 2.2) reacts with

thiolate anions 1500—fold faster than the neutral

133

disulfide. For urease, the equation describing loss of

ureolytic activity by this mechanism would be:
(app :11 [E-S'] [R] + (2 [E-S'] [R W] (8)

with £1 = 105 M‘ls‘l and £2 = 1011 M'lS‘l. This explanation
is implausible for DTDP, since kais faster than the dif—
fusion rate.

A second mechanism consistent with the results in
Figure 5 requires the presence of two essential thiols in
urease, similar to the mechanism proposed for the reaction
of the two active site thiols in thioredoxin reductase
(10). In this model, one urease thiol (with a very high
pkg = 12) reacts rapidly (Q); a second thiol reacts more
slowly (hp and has a low pK} (< 5). The apparent inac-

tivation rate constant is given by:

K1
£1 [H+] + £2
|H+| I
1 + + -—————

K2 [H+]
A key point of this model is that inactivation is caused
by modification of predominantly one thiol at pH values
above 8 and by modification of a second thiol below pH 8.
The uniform effect of boric acid on IAM inactivation at pH
6.0, 7.75, and 9.5 appear to preclude this model. If two

distinct thiols were reacting at different pH values, one

would not expect that a competitive inhibitor would affect

134

the rates equivalently by decreasing the i for modifi—

app
cation of either thiol by a factor of 3.

Our best interpretation of the urease inactivation
results is shown in Scheme 3 (modified from Roberts et

al., 21), which assumes that the essential cysteine

interacts with a second ionizing group (X).

C:

/ \
\<:/.\

inactive enzyme

fl':

inscttve enzyme
(Scheme 3)
Two forms of the enzyme can react with IAM or DTDP with
second-order rate constants indicated by (:1 and £2. The
observed macroscopic equilibrium constants K} and K31 can
be related to the individual microscopic equilibrium con—
stants of Scheme 3 by equations 10 and 11 (Roberts et al.,
1986).
K} = K; + K; (10)

K3 = K11 + K51 (11)

The apparent rate constant for inactivation can then be

represented by

 

Itapp = 1 [H+] + KII (12)
KI [3+]

When the data in Figure 5 were fit to equation 12, values
for pKI, pKII, (gig/KI, and £1 were <5, 12, 0.032 M154, and
230 M484, for IAM, and < 5, 12, 11 M454, and 2.0 X 105
M‘ls‘1 for DTDP. Values for £2 alone could not be deter-
mined because K3 and K4 (Scheme 3) are unknown. Charact-
erization of the individual equilibrium constants would
require additional experiments; ag.potentiometric titra-
tions of these residues could be carried out analogous to
studies involving the active site thiol and histidine of
papain (11). Nevertheless, referring to Scheme 3 and the
results in Figure 5, the ionization state for X must af-
fect the thiol pK such that either pk; approaches 12 when
X is deprotonated or pkg approaches 5 when X is protona-
ted.

An alternative but related model also invokes the
presence of a second ionizing group (X); however, here a
single intermediate form of the enzyme exists in which the
two residues are hydrogen-bonded. The apparent rate con-
stant for inactivation can again be represented by equa-
tion 9.

In summary, the pH dependence of bacterial urease
inactivation is most consistent with a single essential

thiol interacting with a second ionizable residue

136

(together yielding pk; values of < 5 and 12). By con-
trast, Norris and Brocklehurst (19) concluded that the es-
sential thiol in jack bean urease has a pk; = 9.15 and
does not react with another ionizing group. They examined
the essential thiol using DTDP as a Spectrophotometric
probe after first blocking the surface thiols with NEM,
DTDP or DTNB. Although activity measurements were not re—
ported in their pH dependence studies, the spectroscopic
reactivity results yielded no plateau in the low pH end of
the log rate versus pH plot, down to pH 7.0. However, the
jack bean urease study was carried out using phosphate
buffers below pH 8. Figure 4 illustrates that phosphate
decreases the rate of inactivation of the bacterial
enzyme. Furthermore, we have shown (26) that phosphate
inhibition is pH dependent, with phosphate becoming a more
potent inhibitor as the pH decreases. Thus, a low pH
plateau in the jack bean study may have been masked by

phosphate-dependent thiol protection.

Urease inactivation by DTNB- In contrast to the
simple, second-order reaction observed between urease and
six of the reagents examined, the pseudo-first—order
rates of inactivation were not proportional to DTNB con-
centration. Rather, the data were fit to Scheme 2, invol-
ving a preliminary DTNB binding step prior to disulfide
exchange leading to inactivation. This type of saturation

behavior is not uncommon (9); indeed similar kinetics

137

could occur for the other reagents, but the saturating
concentrations are higher than the range examined. The pH
dependence of urease inactivation by DTNB is consistent
with a modified Scheme 3 which includes reversible forma-
tion of a complex between enzyme and modifying reagent.
Binding is unaffected by the ionization status of the es-
sential thiol, whereas the DTNB inactivation step behaves
similarly to amp in urease reactions with IAM or DTDP.

The effect of urea on the rate of urease modification

by DTNB can be accounted for, in part, by a modification

of Scheme 1 to yield

Ki k

 

 

 

E + R \ \ E R -—-‘—> inactive enzyme
+L +L
KL L
Ki k
E L + R \ \ EL R ——> inactive enzyme

 

(Scheme 4)

As shown by Cardemil (1987), the rate of inactivation in

this case is

1R1[KL11 Ki +12 1L1 Ki]

 

kapp : u u (13)
Ki KL (Ki + [R]) + Ki [L] (Ki + [R])

The observed parallel lines in the l/Qmp'versus 1/[DTNB]

plot (Figure 6) require that

138

it1 Ki
it2 K;

so that-the binding affinity increases (Ki decreases) as
the rateefiz decreases with increasing concentration of
ligand. We do not have an explanation for this fortuitous

relationship.

Spectroscopic quantitation of urease thiols- Using
subunit molecular weights and cysteine contents derived
from recently published sequence data (16) and the (1213474
subunit stoichiometry derived form integration of bands
in Coomassie Blue stained SDS gels (25), the holoenzyme
is predicted to possess 20 thiols. Modification of dena-
tured enzyme by DTDP showed that urease possessed 25.5 i
0.3 mol thiol/mol enzyme based on Lowry protein deter-
mination. These results are consistent with an absence
of disulfides in the native protein. Spectroscopic an-
alysis of jack bean urease using DTNB combined with the
amino acid composition indicated that 15 thiols and one
disulfide were present per subunit (Afi.= 96,600; 20),
whereas protein sequencing later demonstrated the pre-
sence of only 15 cysteines per subunit (Nfi.= 90,770; 24).

We have previously shown that K. aerogenes urease
possesses two active sites per native molecule (26).

hence, we can conclude from the spectroscopic analysis

139

that 4 thiols per catalytic unit react with DTDP at pH
7.75. Modification of at least one of the these thiols
is associated with urease inactivation. In the presence
of PPD, a tight-binding competitive inhibitor of urease,
only 3 DTDP reacted per catalytic unit, and the enzyme
was active after PPD removal. These results are consis—
tent with the presence of a single essential thiol per
active site. AHA, a slow binding inhibitor, did not pro—
vide comparable protection of the essential thiol.
Studies at pH 6.1 using the competitive inhibitor phos-
phate also demonstrated that inactivation was associated
with the modification of a single thiol per catalytic

unit.

10.

11.

12.

13.

14.

15.

References

Andrews, A. T., & Reithel, F. J. (1970) Arch.
Biochem. & Biophys. 141, 538-546

Andrews, R. K., Blakeley, R. L., & Zerner, B. (1988)
in The Bioinorganic Chemistry of Nickel (Lancaster,
J. R., Jr, ed) pp. 141-165, VCH Publishers, Inc.,
New York

Bednar, R. A (1990) Biochemistry 29. 3684-3690

Brocklehurst, K., Stuchbury, T., and Malthouse, J. P.
G. (1979) Biochem. J. 183. 233-238

Cardemil, E. (1987) in Chemical.Mbdification of
Enzymes (Jaime Eyzaguirre, ed) PP 23—34, Ellis
Horwood Limited, West Sussex, England

Dixon, N. E., Riddles, P. W., Gazzola, C., Blakeley,

R. L. & Zerner, B. (1980) Can. J. Biochem. 58.
1335—1344

Gorin, G. & Chin, C—C. (1965) Biochim. Biophys. Acta
99. 418-426

Grassetti, D. R., & Murray, J. F. (1967) Arch.
Biochem. Biophys.119. 41-49

Kitz, R. and Wilson, I. B. (1962) J. Biol. Chem.
237. 3245-3249

Kallis, G-B & Holmgren, A. (1980) J. Biol. Chem.,
255, 10261-10265

Lewis, S. D., Johnson, F. A., & Shafer, J. A. (1976)
Biochemistry 15. 5009-5017

Lowry, O. H., Rosebrough, N. J., Farr, A. L. &
Randall, R. J. (1951) J. Biol. Chem.193. 265-275

Mahadevan, S., Sauer, F. D., & Erfle, J. D. (1977)
Biochem J. 163. 495—501

Marquardt, D. W. (1963) J. Soc. Indust. App. Math.
11. 431-441

Mobley, H. L. T. & Hausinger, R. P. (1989)
.Microbiol. Rev. 53. 85-108

140

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Mulrooney, S. M.,

141

(1989) J. Gen. Microbiol. 135. 1769—1766

Mulrooney, S. M.,

172. 5837-5843

Nakano, H., Takenishi, S
Agric. Biol Chem. 48, 1495—1502

Norris, R. & Brocklehurst, K. (1976)

159. 245-257

L.,

Biochem J.,

and

Olson,

Riddles, P. W., Andrews, R. K., Blakeley, R.
Zerner, B. (1983) Biochimica et biophysica Acta,
743. 115-120

Roberts, D. D., Lewis, S. D., Ballou, D. P.,

T. & Shafer, J. A. (1986) Biochemistry 25.

5601

Saada, A.-B. &

Kahane, I.

(1988) Zentralbl.

Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt.

Orig. Reihe A 269,

Schreiner, W.,

(1985) PC Tech. J. 3.

Takishima, K.,
Biochem. 175.

Tsou, C. L. (1962)

Kramer, M.

Suga, T.,
151-165

Todd, M. J. & Hausinger,
Chem. 262. 5963-5967

Todd, M. J., & Hausinger, R. P. (1989)

Chem. 264. 15836-15842

Weatherburn, M.

Winkler, R. G.,
Randall, D. D.
100

Blevins,

160-167

& Mamiya, G. (1988)

1

5595-

Scientia Sinica 11, 1535-1558

R. P. (1987) J. Biol.
J. Biol.
W. (1967) Anal. Chem. 39. 971—974
D. G., Polacco, J. C., &
Trends Biochem. Sci. 13. 97-

(1988)

Pankratz, S. & Hausinger, R. P.
& Hausinger, R. P. (1990) J. Bact.
. & Watanabe, Y. (1984)

S.

, Krischer, S. & Langsam, Y.
vol. 5, 170-190

Eur. J.

IDENTIFICATION OF THE ESSENTIAL CYSTEINE RESIDUE IN
RIJHESIEHHM!.AEHHWSENEIIIHUEASE

.ABSTTUKEP

During reaction with l4C—iodoacetamide at pH 6.3,
radioactivity was incorporated primarily into a single K.
aerogenes urease peptide concomitant with activity loss.
This peptide was protected from modification at pH 6.3 by
inclusion of phosphate, a competitive inhibitor of urease,
which also protected the enzyme from inactivation. At pH
8.5, several peptides were alkylated; however, modification
of one peptide, identical to that modified at pH 6.3,
paralleled activity loss. The N-terminal amino acid
sequence and composition of the peptide containing the
essential thiol was determined. Previous enzyme inactiva-
tion studies of K. aerogenes urease could not distinguish
whether one or two essential thiols were present per active
site (20); we conclude that there is a single essential
thiol present and identify this residue as Cyslm in the

large subunit of the heteropolymeric enzyme.

142

INTRODUCTION

Urease (urea amidohydrolase, EC 3.5.1.5) is a nickel-
containing enzyme that catalyzes the hydrolysis of urea to
form ammonia and carbamate; carbamate spontaneously hydro-
lyzes to H¢CO3 and a second molecule of ammonia. The best
characterized urease is the homohexameric protein from
Canavalia ensiformis (jack bean) which contains 2 nickel per
subunit (NQ.= 90,770) (reviewed by Andrews et al., 2). The
function of urease in this and other plants is unknown (21).
In contrast, bacterial ureases play important roles in ni-
trogen metabolism and have been implicated as virulence
factors in various human and animal diseases (reviewed by
Mobley and Hausinger, 9). The most extensively studied
bacterial urease is that from the Gram-negative enteric
bacterium, Klebsiella aerogenes (currently K. pneumoniae).
This enzyme possesses three different subunits [Nfi.= 60,304
(a), 11,695 ([3), and 11,086 (7)], in an apparent a2fl4‘y4
stoichiometry, and has two bi-nickel active sites (18, 19).
The DNA-sequence of the K. aerogenes urease operon (11) re-
vealed that each bacterial subunit possesses significant ho-
mology to portions of the plant urease sequence, consistent
with gene fusion in the later case. As described below,

both of these enzymes possess an essential cysteine residue;

143

144

however the bacterial and plant enzyme thiols differ in
their chemical properties (13; 20).

Labelling studies of jack bean urease indicated that 4
cysteines/subunit react rapidly with a disulfide reagent,
DTDP, with no loss of activity, whereas modification of a
fifth cysteine was correlated to activity loss (13). The
essential thiol exhibited a pH dependence of reactivity
similar to that of a free thiol, with a pKh = 9.1. Taki—
shima et al. (16) followed up on this work by modifying the
non—essential enzyme thiols with N-ethylmaleimide, then
derivatizing the essential thiol with the chromophore N-(4—
dimethylaminodinitrophenyl) maleimide. They demonstrated
the specific labelling of a single cysteine residue, Cyssmp
In contrast, Sakaguchi et al. (15) partially modified jack
bean urease with diazonium-l H-tetrazole, then added 1%}-
1abelled reagent to derivatize the more slowly reacting
thiols. They showed that label was incorporated into 2
cyanogen bromide fragments; one fragment was further studied
and contained Cysmn. More recently, jack bean urease was
modified using N-iodoacetyl-N'(5—sulfo-l—naphthyl)ethylene
diamine and N-(7-dimethylamino-4—methyl-3—coumarinyl)male-
imide (17): three thiols were modified with no loss of ac-
tivity, whereas alkylation of 2 additional thiols (Cyszv7
and Cysan) was concomitant with activity loss. Complete
modification of Cysun with the latter reagent resulted in
an activity decrease of only 50%; thus the authors concluded

that CYSgn (but not Cysfln) is essential for activity.

145

No active site labelling studies have been reported for
any bacterial urease; however kinetic analysis of K.
aerogenes urease revealed a pseudo-first-order loss of ac-
tivity in the presence of disulfide reagents or alkylating
reagents (20). The addition of substrate or inhibitors
altered the rates of inactivation, consistent with local-
ization of the essential thiol proximal to the active site.
Competitive inhibitors (PPD at pH 7.5, and phosphate at pH
6.1) were shown to protect one thiol per active site from
modification by DTDP. The apparent second-order rate con-
stant for activity loss as a function of pH (using either
IAM or DTDP) differed notably from the behavior of the plant
enzyme in being constant over the range of pH 5 to pH 8.
The results are consistent either with the essential thiol
interacting with a nearby ionizable amino acid or with the
presence of two essential cysteines per active unit (20).
In the latter case, inactivation at low pH would involve
modification of a cysteine which is distinct from that
modified at high pH; both cysteine residues would be
required for activity.

This manuscript describes our efforts to identify the
essential thiol(s) of K. aerogenes urease and to discrimi-

nate between the one- and two-cysteine models.

EXPERIMENTAL PROCEDURES

Enzyme Purification— K. aerogenes urease was purified
to a specific activity >2500 umol-min*bn@“1 from strain
CG253 carrying the recombinant plasmid pKAU19 (12) by using
procedures previously described (19). The third and final
step of the purification, Mono-Q (Pharmacia) anion exchange
chromatography, made use of DTT-containing buffers (1 mM) to
ensure the absence of mixed-disulfides between urease and B-
ME. Enzyme was stored at O’C in DTT-containing buffers un-
til used. Protein was assayed by the method of Lowry et al.
(8) using bovine serum albumin as the standard and confirmed
using the extinction coefficient 8:30“ = 8.5 (18) . The mo-
lecular weight of urease was calculated to be 211.8 ug/nmol

(11).

Uniform labelling of cysteine residues in urease-
Purified urease was desalted by ultrafiltration using a Cen-
tricon-30 microconcentrator (5 X 1:10 dilution) to remove
thiols used to stabilize the enzyme during storage, then
dried using a speed-vac concentrator (Savant). 15 mg urease
was denatured in 6 M Gn-HCl, 10 mM EDTA, 0.1 M Tris-HCl, pH
8.0, reduced by incubation with DTT (500 nmole) for 60 min
at 50°C, and carboxamidomethylated with 14C-IAM (2.5 mole @
0.185 GBq/mmol, Amersham) for 30 min at 25°C. The modifica-

tion reaction was quenched by adding excess B-ME.

146

147

Specific labelling of the cysteine residues in native
urease- K. aerogenes urease (20-50 nM) was desalted into 1
mM EDTA, 80 mM buffer (MES, pH 6.3, HEPES, pH 7.75, or TAPS,
pH 8.5) and allowed to react with 5 mM 1NZ-IAM at 37°C until
< 5% of the activity remained, as determined by assaying
aliquots using a standard assay procedure (19). Reactions
at pH 6.3 were done both with and without 20 mM phosphate, a
pH-dependent competitive inhibitor of urease (19). to pro-
tect the essential cysteine from modification. Some reac—
tions at pH 7.75 included the competitive inhibitors AHA (25
mM) or PPD (25 nM). During the reactions, aliquots were
quenched by addition of excess B-ME, then desalted on a
Superose-12 gel filtration column (1 X 30 cm, Pharmacia),
equilibrated with 1 mM EDTA, 20 mM phosphate, pH 6.5, con-
taining 1 mM B-ME. Radioactively labelled urease samples
were denatured in 6 M Gn-HCl, reduced with DTT, and the re-
maining cysteines were modified with unlabeled IAM using

procedures analogous to those described above.

Standardized protocol for peptide mapping- The a sub-
unit of urease was separated from the B and Y subunits by
Superose-12 gel filtration chromatography (1.0 X 30 cm) in 3
M Gn-HCl, 1 mM EDTA, 0.1 M Tris-HCl, pH 8.0. Fractions con-
taining the a subunit were pooled, concentrated, and ad-

justed to 1 M Gn-HCl. CaClZ was added to 1.0 mM, then

148

samples were digested with 4% TPCK-treated trypsin (TYpe
XIII, Sigma) for 24-40 hours at 37'C (2% added initially,
another 2% added after 8—12 hours digest). Tryptic peptides
were separated by using reverse phase chromatography (Pro-
RPC, 0.5 X 10.0 cm, Pharmacia), with a multistep gradient of
0—50% IPN containing 0.05% TFA. Two peptides which did not
bind to the Pro—RPC could be separated on a Pep-RPC column
(Pharmacia) with a 0-100% acetonitrile gradient, containing

0.1% TFA.

Characterization of the specifically labelled peptide-
Sequence analysis of the specifically-labelled tryptic pep-
tide was done by using an Applied Biosystems Model 477 pro-
tein sequencer followed by HPLC analysis of phenylisothio—
cyanate derivatives. Amino acid composition was determined
by HPLC analysis after converting the hydrolyzed sample to

the phenylisothiocyanate derivatives.

RESULTS AND DISCUSSION

Uniform labelling of urease cysteines and development
of a peptide mapping protocol- Denatured K. aerogenes
urease was uniformly labelled with 14C-IAM at pH 8.0. The a
subunit (M1. = 60,304) was separated from the B and 7 sub-
units (Afl.= 11,695 and 11,086, respectively) by gel-filtra—

tion chromatography (Figure 1), which also removed excess

149

 

 

 

 

 

I Tl ' ' I I T I F I t r r T
5000:
30001
cpm I
10005 I I
j __ .---____I -_
L==__JL..=_J L I
a 5+7
Excess
Reagent
0.4—
280nm
0.2a
0.0 ,....r,...?..A/{\.:~.,
5 to 15 20 25

Vdunm OnD

Figure l. Subunit distribution of urease thiols.
Urease was denatured, reduced, and uniformly modified with
l4C-IAM. Alkylated protein was applied to a Superose-12
column (1.0 X 30 cm, Pharmacia) equilibrated with 3.0 M
Gn~HCl, 10 mM EDTA, 0.1 M Tris-HCl, pH 8.0. Cpm was
determined in 5% of each 1.0 ml fraction, and absorbance was
monitored at 280 nm. Brackets indicate fractions included
in calculations of the relative amount of label in each

subunit.

150

reagent. The ratio of 14C in the a subunit to that in the
sum.of the B and 7 subunits was 4.8:1. The B and 7 subunits
could be separated by reverse phase chromatography (data not
shown), and all of the radioactivity was found in the B sub-
unit; thus the ratio of thiols in the a, B, and 7 subunits
was 4.8:lzo. The genes encoding these proteins (11) indicate
an 8:1:0 ratio of thiols in the a, B, and 7 subunits.

Hence, there are twice as many B as a subunits in the native
enzyme, consistent with the proposed a2B4y4 subunit structure
(18).

Purified carboxamidomethylated-urease a subunit was
highly insoluble unless maintained in the presence of a de-
naturant such as l M Gn-HCl. A standard protocol for pep-
tide mapping was developed by digesting the subunit dis—
solved in this chaotropic agent with TPCK-treated trypsin
for various lengths of time and analyzing the peptides by
Pro-RPC chromatography (Figure 2). Optimum resolution of
cysteine-containing peptides (determined by monitoring ab-
sorbance at 214 nm and by analysis of fractions for cys-
teine-containing peptides using liquid scintillation coun-
ting) was only achieved after 30—40 hours of digestion, at
37'C. Six major cysteine-containing peaks were identified
as flow-thru (FT), and peptides A, B, C, D, and E (Figure
2). The flow-thru fraction could be further resolved into
two additional peptides, F and G, on the more hydrophobic
column, Pep—RPC (data not shown). This protocol yielded

consistent peptide maps in which the 7 peaks accounted for

151

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

800- -
W
500 A 1:E 50
C 5: F
D i _
Cpm) 400 . 3 40
200 V E (30
% IPN
0-44 o...---‘~2o
o.3« .
Absorbonce ’,'
214 nm 0.2- .10
0.1. W“ MN
0.0 .. W,
o 50 60

Volume (ml)

Figure 2. Pro-RFC analysis of uniformly-labelled
trypsin-digested urease arsubunit. Purified urease a-

subunit was digested with TPCK-treated trypsin (as described
in the text) and applied to a Pro-RPC column.

The column
was eluted using the indicated gradient of IPN/0.05 % TFA

(dotted line) while monitoring the absorbance at 214 nm
(solid line).

Fractions (1 ml) were collected and radio-

activity was determined in 0.1 ml aliquots by scintillation
counting (dashed line).

152

at least 75 % of the applied label in uniformly-labelled
samples, and clearly resolved the specifically labelled pep-

tides in the experiments described below.

Specific labelling of the essential cysteine at pH 6.3—
The reactivity of the essential thiol of K. aerogenes
urease exhibited a pk; of 12 and decreased below this value
until about pH 8; below pH 8 the essential thiol showed a
constant rate of modification (20). In contrast, the reac-
tivity of most thiols towards modifying reagents decreases
with decreasing pH below the thiol pkg. We attempted to
exploit this low pH reactivity as a method to specifically
label the essential thiol. In addition, since the competi-
tive inhibitor phosphate was shown to completely protect
urease from inactivation by both disulfide-forming reagents
and alkylating agents (20), we sought to identify the essen—
tial cysteine modified at pH 6.3 by comparing the peptides
labelled by lflt-IAM in the presence and absence of phos-
phate. After 25 hours incubation with L4C-IAM in the ab-
sence of phosphate, a total of 7.7 thiols (6 in the a sub-
units, l.7 in the B-subunits) were modified with complete
loss of activity; in the presence of phosphate, 4 thiols in
the a subunits were protected from modification (Figure 3).
Phosphate did not protect any cysteine in the B subunit,
demonstrating that the essential cysteine is in the a sub-
unit. The finding that only 2 B-subunit thiols are modified

in inactivated enzyme may indicate that the 4 B-subunits are

 

   
 
     
    
  

1 o I U V I I I V l T I I
_____ o- _ J
0.3- .
Fraction .
O O 006‘ C1
Actwnty

Remaining 0.4-

4

(L2-

 

ITtIITTTWIT‘rTIIIYTIYUTT'

1

65%

5.0-
Number of .

Thiols ‘

Rea cted 3'0:
2.0-

 

 

 
 

 

 

VT—TITrTTII" YT' I'l'i"

o 5 10 '1'5‘ 20 25
Time (hours)

Figure 3. Effect of phosphate on the modification of
urease thiols by IAM. Urease was inactivated with 5 mM LR}-
IAM in the presence (open symbols) and absence (closed
symbols) of 20 mM phosphate, in 1 mM EDTA, 80 mM MES, pH
6.3. A) The remaining activity is shown as a function of
time. B) The number of a-subunit (squares) and B-subunit
(triangles) thiols reacted was assessed as described in the

text.

154

present in 2 distinct conformations: the thiols in two of
these subunits are accessible to solvent, whereas the other

2 B-subunit thiols are sequestered.

Purified a—subunit was digested with trypsin, and sub-

jected to the Pro-RPC peptide mapping protocol. In the ab-

sence of phosphate, the label was incorporated most specifi-
cally into peptide D, and the extent of peptide modification
was reasonably well correlated to activity loss. Further-

more, phosphate provided significant protection of this same
cysteine-containing peptide (Figure 4). Other cysteine-con—
taining peptides were protected to a lesser extent, probably
because phosphate stabilizes the enzyme from low-pH induced

denaturation.

Specific labelling of the essential cysteine at high
pH- At higher pH values, competitive inhibitors of urease
did not protect the essential thiol from alkylation by IAM
(Figure 5). This result was surprising because we had pre-
viously shown that the slow-tight binding inhibitor PPD (K}
= 95 pM), but not AHA (K}=:2.6 pM), protected 1 mole of
thiol per mole of active site from modification by DTDP
(20). PPD may be able to sterically exclude DTDP from the
active site, whereas the smaller, uncharged alkylating re—
agent, IAM, is still able to react with the essential thiol.

Because the competitive inhibitor protection approach

was untenable for identification of the essential cysteine

155

 

 

 

 

 

‘LO
7‘
(LB-
3 ‘ 10hr “hr 2°”
3‘3"? o 6-
‘00 '
OI
E Q 7
U)
3§15 Ouk- ll
5 I
4—1 d y
5535" , 355'
. SSN ’5 . I :I'IN :II
'i'1°§ 7 {4 ’//I .;.;.§ .;.;.I
o 0 25:3 1:1 %. E125! 52:38) 32:3 k? 1:39 Izt

 

Figure 4. Peptide distribution of urease thiols
modified at low pH. The a-subunit of urease, modified at
pH 6.3, was purified and digested with trypsin. Pro—RPC
chromatography was used to determine the amount of labelled
thiol in each peptide in the presence (+) and absence (-) of

phosphate.

156

 

Y T IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 8
l I I I
d
)-
1004?
. \
‘\

l e

d “ .
ao— ‘ *5

.

 

 

 

'o
0.)
.4 “ E
§ " 'o
eo~ \ ,//‘ ~ g
.. ‘v
% ActIVIt « \ c4
y ~ 1’
remamm ‘ a a. 9
9 40— *x -c
. / cg _ r-
q “\..~‘ _ 2 %
20"I “~‘.‘.. )-
........... .---- u-
- ________ .u
0 IIIIIIIII I I rrrrrrrrr T T TTTTT I ''''''' T T r 1 WT T T I F 0
O 5 10 15 20 25

Tune (hours)

Figure 5. Effect of competitive inhibitors on urease
thiol modification at pH 7.75. Urease was inactivated at pH
7.75 in 10 mM EDTA, 100 mM HEPES, pH 7.75 using 14C-IAM, in
the absence of inhibitors (0), and in the presence of 25 mM
AHA (I) or 25 MM PPD (A) . Aliquots were quenched with an
excess of B-ME, desalted using Superose-12 gel-filtration
chromatography, and the incorporated radioactivity was
measured to determine the number of thiols modified. The
time-dependent decrease in activity is that for urease
inactivated in the absence of inhibitors (4»). Activity of
the inhibited samples was not determined because the rate of
inhibitor dissociation to yield active enzyme is slow (T05

~ 2.5 hr, Todd and Hausinger, 1989).

157

residue at higher pH values, the reactivity of individual
cysteine-containing peptides was examined in uninhibited
samples as a function of remaining activity. The pattern of
subunit labelling at pH 8.5 is shown in Figure 6a. 2-3
thiols were modified with essentially no loss of activity;
activity loss occurred upon modification of an additional 5-
6 thiols per halo-enzyme, with 1-2 thiols modified in the B
subunits, and 4 thiols modified in the a subunits. Tryptic
peptide analysis of a-subunit purified from urease modified
at pH 8.5 revealed that activity loss paralleled the
labelling of only peptide D (Figure 6b).

As in the pH 6.3 study, only half of the B-subunits
were alkylated at pH 8.5, perhaps indicating that the halo-
enzyme has two distinct populations of B-subunits. (We
assume that the same two thiols are reactive at each pH).
The low pH/inhibitor protection analysis demonstrated that
the B-subunit thiol is not essential for catalysis; however,
modification of this thiol (at pH 8.5) closely followed ac-
tivity loss in the absence of inhibitors. The observed
similarity in reaction rates between the B-subunit thiol and
the essential thiol in K. aerogenes urease may relate to the
anomalous jack bean urease active site labeling studies of
Sakaguchi et al., (15). The bacterial B-subunit is homolo-
gous to residues 132-237 in the jack bean urease sequence

(11, 17) and the only cysteine in this region of the plant

urease is Cysmfl; Sakaguchi et al. (15) may have

158

 

 

 

 

 

4 4
O.9~ .
. 0.7- -
Fraction '
Actwuiy. 0.5« J
Remaining ‘
a3~ .
i
1 «i
OJ- .
to a
.. ‘\ 5 i «'5
‘ I.A
o.s- ‘ B
e C
4 ' D
. a
Fraction 0-6- . E + G
Acfivhy i \
Remaining 0.4- K
\
QO . , . . . , .
an 02. 0A- at as

Thiols modified

Figure 6. Peptide distribution of urease thiols
modified at high pH. Urease was inactivated with 5 mM 1%}—
IAM at pH 8.5. A) The activity remaining as a function of
the total number of thiols modified (0) was determined. In
addition, the distribution of label in the a (I) and B (A)
subunits was measured. B) Activity is shown as a function

of the amount of label incorporated into the a-subunit

tryptic peptides.

159

misidentified Cysun as the essential cysteine if this
residue reacts similarly to that in the bacterial B-subunit.
In contrast to jack bean studies reporting the rapid
modification of several thiols followed by slower reaction
with the essential thiol (13, 1516, 17), we saw no cysteine-
containing peptide in the a-subunit labelled faster than
peptide D. Rather, our peptide mapping protocol demonstra-
ted that a small amount of each cysteine-containing peptide
was labelled with no loss of activity, perhaps due to the

presence of a small (<10%) fraction of inactive protein.

Identification of the cysteine in peptide D- N-
terminal sequence analysis of peptide D, purified from
uniformly labelled urease, revealed the sequence TLNTI.
Comparison of this sequence to that predicted from DNA
sequence analysis (11) allows us to conclude that peptide D
begins at residue 305 of the a—subunit. This peptide is
produced by cleavage after Tyrum, indicating that after 30
hours digestion, significant Whtryptic cleavage had oc-
curred. WLtrypsin is an autolytic product of trypsin which
possesses chymotryptic-like activity, although at a much
slower rate (1). Peptide D was subjected to amino-acid
analysis (Table l). The composition is consistent with
peptide D extending from position 304 to the tryptic
cleavage site at Arglm. The only cysteine in this peptide

is Cysym. We conclude that K; aerogenes urease possesses a

160

Table 1

Amino Acid Composition of the Essential Cysteine-Containing

Peptide

 

Amino Predicted Experimental
Acid

 

Asx
Glx
Ser
Gly
His
Arg
Thr
Ala
Pro
Cys1
Tyr
Val
Met
Val
Ile
Leu
Phe
Lys
Trp

\IKO

OHpNOONOOHWNHNNHb-itb

:3
Q0 0 o o
NprpomH(nmpp(pr-Jp\lii.

OOHQNONNI—‘l—‘HWNHWOHWQ

 

1 Quantitated as carboxymethyl cysteine (although modifica-
tion reactions were carried out using IAM, the carboxam-
idomethyl derivative is hydrolyzed to the carboxymethyl
derivative during acid hydrolysis in 6 N HCl).

2 not detected

161

single essential cysteine residue, and identify this residue

as Cyslm.

Comparison of K. aerogenes urease sequence to other
ureases- Ureases from K. aerogenes (ll), jack bean (17),
Helicobacter pylori (4, 6), Ureaplasma urealyticum (3),
Proteus mirabilis (5), and Proteus vulgaris (10) have been
sequenced. Only one cysteine (Cysxm by K; aerogenes num-
bering or Cysan by jack bean numbering) is conserved among
all species. As shown in Figure 7, the sequence immediately
surrounding the essential cysteine, is identical at 22 of 29
amino acids. This region also contains three conserved his—
tidine residues at 312, 320 and 321; histidines have been
suggested to participate in binding the active site nickel
ions (14, 7). Additionally, one of these histidines could
interact with the essential thiol to enhance its reactivity
at low pH values (20). Although data from both the bac-
terial and plant urease studies are consistent in iden—
tifying the essential cysteine, its precise role in

catalysis remains unknown.

162

** *‘k'k'k'k * * *‘k ** *** ****** *‘k ** **

Jb 560-VCGIKNVLPS STNPTRPLTS NTIDEHLDML MVCHHLDREI PEDLAFAHSR-610
Hp 289-VAGEHNILPA STNPTIPFTV NTEAEHMDML MVCHHLDKSI KEDVQFADSR-339
Uu 293-SVKYAHILPA STNPTIPYTV NTIAEHLDML MVCHHLNPKV PEDVAFADSR-343
PV 287—SVGEPNILPA STNPTMPYTI NTVDEHLDML MVCHHLDPSI PEDVAFAESR-337
Pm 287-SVGEPNILPA STNPTMPYTI NTVDEHLDML MVCHHLDPSI PEDVAFADSR-337
Ka 287-ACAHPNILPS STNPTLPYTL NTIDEKLDML MVCHHLDPDI AEDVAFAESR-337

Figure 7. Comparison of urease sequences surrounding
the essential cysteine residue. The two letter abbrevia-
tions stand for: Jb-jack bean (Takishima et al., 1988);
Hp-Helicobacter pylori (Labigne et al., 1991); Uu-Ureaplasma
urealyticum (A. Blanchard, 1990); Pv-Proteus vulgaris
(Morsdorf & Kaltwasser, 1990); Pm-Proteus mirabilis (Jones &
Mobley, 1989); Ka-Klebsiella aerogenes (Mulrooney &
Hausinger, 1990). The indicated amino acid positions refer
to the a-subunits of the bacterial ureases. Residues which
are identical among all sequences are marked by an asterisk;
the conserved cysteine and the three conserved histidines

are highlighted in bold.

10.

ll.

12.

13.

14.

15.

REFERENCES

Andrews, R. K., Blakeley, R. L. & Zerner. B. (1988) in
The Bioinorganic Chemistry of Nickel (Lancaster, J. R.,
Jr ed) pp. 141—165, VCH Publishers, Inc., New York

Allen, G. (1981) in Laboratory Techniques in
Biochemistry and Mblecular Biology. (T. S. Work & R. H.
Burdon, eds) pp. 52-58, North Holland Pub. Co., NY.

Blanchard, A. (1990) Molec. Microbiol. 4, 669—676
Clayton, C. L., Pallen, M. J., Kleanthous, H., Wren, B.
W., & Tabaqchali, S. (1990) NUcleic Acids Res. 18,
362

Jones, B. D. & Mobley, H. L. T. (1989) J. Bacteriol.
171, 6414-6422

Labigne, A., Cussac, V. & Courcoux, P. (1991) J.
Bacteriol. 173, 1920-1931

Lee, M. H., Mulrooney, S. B. & Hausinger, R. P. (1990)
J. Bacteriol. 172, 4427-4431

Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall,
R. J. (1951) J. Biol. Chem. 193, 265-275

Mobley, H. L. T., and Hausinger, R. P. (1989)
Microbiol. Reviews 53. 85-108.

Marsdorf, G. & Kaltwasser, H. (1990) FEMS.Micro. Let.
66, 67-74

Mulrooney, S. B. & Hausinger, R. P. (1990) J.

Bacteriol. 172, 5837-5843

Mulrooney, S. B., Pankratz, H. S. & Hausinger, R. P.
(1989) J. Gen Microbiol. 135. 1769-1766

Norris, R. & Brocklehurst, K. (1976) Biochem. J. 159,
245—257.

Sakaguchi, K., Mitsui, K., Kobashi, K., & Hase, J.
(1983) J. Biochem. 93. 681-686

Sakaguchi, K., Mitsui, K., Nakai, N. & Kobashi, K.
(1984) J. Biochem. 96. 73—79

163

l6.

17.

18.

19.

20.

21.

164

Takishima, K., Mamiya, G. & Hata, M. (1983) in
Frontiers in Biochemical and Biophysical Studies of
Proteins and Membranes (T. Y. Liu, S. Sakakibara, A. N.
Schecter, K. Yagi, H. Yajima, and K. T. Yasunobu, eds.)

pp. 193—201, Elsevier, N.Y.

Takishima, K,., Suga, T. & Mamiya, G. (1988) Eur. J.
Biochem. 175. 151-165

Todd, M. J. & Hausinger, R. P. (1987) J. Biol. Chem.
262. 5963-5967

Todd, M. J. & Hausinger, R. P. (1989) J. Biol. Chem.
264. 15835-15842

Todd, M. J. & Hausinger, R. P. (1991) J. Biol. Chem.
(in press).

Winkler, R. G., Blevins, D. G., Polacco, J. C. &
Randall, D. D. (1988) Trends Biochem. Sci. 13. 97-100

SUMMARY and FUTURE DIRECTIONS

Upon initiation of this project little was known about
microbial ureases. Most research on this medically and ag-
riculturally important enzyme was done by scientists
studying mixed microbial populations, whole cells, or impure
proteins. The few known characteristics of microbial
ureases were similar to the extensively studied plant
enzyme, thus some researchers exploited the plant enzyme to
draw conclusions about bacterial ureases! This chapter
summarizes the purification and characterization of urease
from K. aerogenes, describes active site studies using
competitive inhibitors and amino acid modifying reagents,

and exposes interesting questions arising from this work.

Urease purification and characterization- Urease of
the Gram negative enteric bacterium, K. aerogenes, was
purified to homogeneity and was demonstrated to have an
a2B4‘y4 subunit structure, containing 4 nickel (chapter 2) .
This was not expected since the well-studied plant enzyme
has a homo-hexameric structure containing 12 nickel
dispersed among 6 active sites. Using the tight-binding
inhibitor PPD we demonstrated that K. aerogenes urease has

two active sites (Chapter 3); therefore both enzymes have 2

165

166

mol nickel/mol active unit. The “25474 subunit stoichio—
metry has not been definitively determined since quan-
titation assumes each subunit has similar binding affinity
for dye; however the presence of three distinct subunits was
confirmed through the DNA sequencing efforts of Scott
Mulrooney. The three subunits share significant homology to
the jack bean enzyme and are consistent with gene-fusion in
the latter case. Gel filtration chromatography in the
presence of salts, glycols, DMSO, reductants, and detergents
failed to cause subunit dissociation (Chapter 2). Perhaps
moderate concentrations of chaotropic reagents (ag.2.0 M
guanidine-HCl) will be found to disrupt subunit
interactions, allowing investigation of the forces
determining quaternary structure and identification of the
nickel binding subunit(s).

Not only are the subunits held together tightly, but
the nickel is also tightly bound. Nickel is released at pH
values below 4.5 and examination of the holoenzyme treated
at low pH may reveal increased reactivity of the nickel
ligands. Low pH treatment may also lead to replacement of
one or both of the nickel with other metals, allowing
further spectroscopic and biophysical characterization.
Nothing, however, is known of the stability of nickel-
depleted urease apo-enzyme to low pH; a low-pH treatment
could "irreversibly" denature the protein.

Purified urease can migrate as several distinct, active

isozymes through native gel electrophoresis. SDS PAGE of

167

 

 

 

2.0— . r4000 [
A 7 “ *4
d L5~ F i
A « 1 T.,, n
E .1 “ .4; E '-3 “IE,
\ ._
o H o- >\
s. U l N
g 1.0— 9 ~2ooo < E g
.9 E ~2‘\
.E r: _. a
Q) 0 O x
+1 0 E U
a. as— L m _1
“ L
0.0 i r 0 b0

 

(o—

14

Fraction

Figure 1. Separation of urease isozymes on Mono-Q.
Urease was applied to a 1.0 x 10 cm Mono-Q equilibrated with
1 mM EDTA, 20 mM phosphate, pH 6.5, containing 1 mM BtME.
Urease protein was eluted with an increasing gradient of
KCl. Protein, activity, and nickel were determined as

described in Chapter 2.

168

each active band showed identical subunit stoichiometry, as
assessed by scanning densitometry. These isozymes can also
be separated by either ion exchange chromatography or
gradient hydrophobic chromatography. The major active form
(isozyme B) migrates faster under native gel electrophoresis
and elutes later from anion exchange chromatography,
consistent with a greater anionic character. One possible
explanation for the multiple forms is that the minor form
has less than 4 mol nickel/mol enzyme. Figure l is a
preliminary experiment determining specific activity,
protein, and nickel content of urease eluting from Mono-Q
anion exchange chromatography. The isozyme which elutes
first (A) has only 75% of the nickel (and 75% of the
specific activity) of the isozyme which elutes later (B)
consistent with A having less than stoichiometric amounts of
nickel. While isozyme B is very stable, it will (over time)
form some isozyme A; isozyme A looses activity quickly at
low pH and has never been observed to form isozyme B. To
simplify experiments, all inhibitor and active site studies
in chapters 3-5 (and all biophysical characterizations)
employed the second isozyme only.

Interpretation of urease structural studies would be
easier with knowledge of the urease crystal structure.
Methods to crystallize plant ureases have been known for 65
years, yet poor crystal quality and large unit cell have
precluded solving the crystal structure. Crystals of the K.

aerogenes enzyme have been obtained (A. Karplus, in

169

progress) and structural studies are continuing.
Integration of crystal structure with chemical and genetic
studies should contribute to understanding the mechanism of

urea hydrolysis.

Urease inhibitors— The intelligent design of useful
urease inhibitors requires detailed knowledge of enzyme—
inhibitor interactions. AHA and PPD were found to be slow
binding competitive inhibitors of K. aerogenes urease
(Chapter 3). A slowly processed substrate is kinetically
indistinguishable from a slow-binding inhibitor, therefore
assays for PPD hydrolysis should be done. If PPD is hydro—
lyzed, the rate of hydrolysis may correspond to the rate of
formation of enzyme-inhibitor complex, or to the rate of
enzyme reactivation. A thorough examination of the pH—
dependence for PPD inhibition of urease may help elucidate
the mechanism of urea hydrolysis. Preliminary studies using
AHA-urease indicated that the rate of reactivation (My
increased below pH 6.5, perhaps indicating the enzyme pl; of
6.55 (required to be deprotonated for urea catalysis) must
be deprotonated for the formation of a tight binding
transition state complex with AHA or PPD.

With the advent of molecular biology and the develop-
ment of expression vectors producing large quantities of
protein, biophysical and spectroscopic studies on the urease
nickel center are now possible. Magnetic susceptibility and

MCD measurements on urease are consistent with two nickel

170

per active site being antiferromagnetically coupled in the
presence of B—ME. Similar studies using the slow-binding
inhibitors PPD and AHA may confirm the proposed bridging
interactions of these competitive inhibitors with urease.
Additional evidence linking the interactions of AHA and PPD
with the nickel center may be deduced if the UV-visible
spectrum of K. aerogenes urease changes upon binding of
these inhibitors. The rate of spectral change may
correspond to the rapid equilibration between enzyme and AHA
or the slow step leading to our proposed "bridging"

intermediate.

Urease active site amino acids- Ureases from all
sources examined are susceptible to inactivation by thiol
specific reagents. Competitive inhibitors modified the
reactivity of an essential thiol in K. aerogenes urease and
protected one thiol per catalytic unit from modification
(Chapter 4). The essential cysteine of the jack bean enzyme
is thought to act as a general acid, donating a hydrogen to
the ammonia molecule as it leaves. We have demonstrated
that at catalytically active pH values, this essential
cysteine has appreciable nucleophilic character and that it
interacts with another ionizable enzyme group (Chapter 4).
The identity of this second ionizable group is not known,
however the model of Zerner predicts three additional ion-

izable groups in close proximity to the active site thiol:

171

a carboxylate, an un—identified base, and a nickel—bound
hydroxide. .

Only indirect evidence for a carboxyl group at the
active site of urease has been published: comparison of R;
values for various nickel binding inhibitors (Chapter 3) and
(app values for several thiol modifying reagents (Chapter 4)
were consistent with repulsion of negatively charged
molecules by the urease active site. Preliminary studies of
K. aerogenes urease showed no loss of activity upon exposure
to Woodwards Reagent K, at 37'C. However this reagent
decomposes rapidly. More thorough analysis using carbo—
diimides or trialkyloxonium salts are obviously needed.
Prior protection of the essential thiol with a reversible
modifying reagent (ag.nmdification by disulfide reagents is
reversible) may be necessary.

Other amino acid specific reagents may be used to
identify amino acids which participate in catalytic
turnover. For example, DEP has been demonstrated to quickly
inactivate urease holoenzyme (Mann Hyung Lee), consistent
with a histidine being essential for activity; however the
effect of active site ligands and the identity of the amino
acid modified is not known. Tetranitromethane could be used
to modify tyrosine side chains; lysine could be guanidinated
using O-methyl isourea or reductively methylated with
formate/NaCNBH4. Preliminary experiments using the latter
two lysine modifying reagents were hampered because the

reagents interfered with the indophenol assay. Figure 2

172

 

 

 

U 5' 6’ c o 4_e
( . Control
. . ' D
_2_ . .
A . o 2.4-pentonedlone
>° .
\_ 1
3 _4_
E ° 2,3—butonedione
_6_.
phenylglyoxol
T T T r r I I T T r T T T T r T T ¢ T r T T T T I I I I
O 25 50 75 100 125

Thne (nfinutes)

Figure 2. Pseudo-first order inactivation of urease by
arginine modifying reagents. Urease was incubated at 37'C
in 8 mM EDTA, 80 mM HEPES, pH 7.75 alone, or in the presence
of 200 mM 2,4—pentane dione, 50 mM 2,3-butane dione, or 100
mM phenylglyoxal. At the indicated intervals, a small
aliquot was diluted into the standard assay mixture to

determine activity.

173

demonstrates the pseudo—first order loss of K. aerogenes
urease activity by three arginine modifying reagents: 2,4—
pentanedione, 2,3-butanedione, and phenylglyoxal. A
comprehensive analysis of the effects of arginine modifying
reagents will include determination of: the apparent
second-order rate constants (and the effect of pH on these
rate constants), the effects of active site ligands, and the
effect of prior modification of the essential thiol.
Identification of the essential arginine residue would
naturally follow.

Clearly, expanded investigation of essential amino
acids could contribute to our understanding of the enzyme
mechanism” Once an amino acid is identified (as Cys319 was
identified as essential for K. aerogenes urease activity;
Chapter 5), site directed mutagenesis, (varying amino acid
size, charge, polarity, or ability to form hydrogen bonds)
could distinguish the properties of the amino acid involved
in catalysis. Similar site directed mutagenesis studies
could be targeted towards proposed nickel ligands.
Simultaneous biophysical, spectroscopic, chemical, kinetic,
and genetic analysis of the urease active site should
generate complementary data, creating a thorough
understanding of the catalytic mechanism and leading to the
rational design of potent inhibitors for this medically and

agriculturally important enzyme.

APPENDIX

SATURATION MAGNETIZATION OF THE NICKEL CENTERS IN UREASE.

E,P, Day, J. Peterson+, M.J. Todd*, and R.P. Hausinger*
University of Minnesota, Minneapolis, MN 55455. +University
of Alabama, Tuscaloosa, AL 35487. *Michigan State
University, East Lansing, MI 48824.

The two nickel ions at the active site of urease are
believed to be in the Ni(II) oxidation state. We have
measured the saturation magnetization of urease isolated
from Klebsiella aerogenes to determine its spin, spin
concentration, average g value, and zero-field splitting.
Activity was measured before and after the magnetization
data were collected. EPR was used to quantitate Fe(III)
impurities and metal analysis to quantitate the nickel.

Spin S=l paramagnetism accounted for roughly half the nickel
implying that half the nickel is low spin (8:0). No
evidence for exchange coupling between the nickel ions was

found.

174